Geobiology (2007),

5

, 351–373 DOI: 10.1111/j.1472-4669.2007.00125.x

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

351

Blackwell Publishing Ltd

ORIGINAL ARTICLES

Emergence of animal life

The Precambrian emergence of animal life: a geobiological perspective

E. GAIDOS,

1,4

T. DUBUC,

2

M. DUNFORD,

2

P. MCANDREW,

3

J . PADILLA-GAMIÑO,

3

B. STUDER,

1

K. WEERSING

3

AND S. STANLEY

1

1

Department of Geology and Geophysics,

2

Department of Zoology, and

3

Department of Oceanography, University of Hawai’i at M

å

noa, Honolulu, HI 96822, USA,

4

NASA Astrobiology Institute

ABSTRACT

The earliest record of animals (Metazoa) consists of trace and body fossils restricted to the last 35 Myr of thePrecambrian. It has been proposed that animals arose much earlier and underwent significant evolution as acryptic fauna; however, the need for any unrecorded prelude of significant duration has been disputed. In thiscontext, we consider recent published research on the nature and chronology of the earliest fossil record ofmetazoans and on the molecular-based analysis that yielded older dates for the appearance of major animalgroups. We review recent work on the climatic, geochemical, and ecological events that preceded animal fossilsand consider their portent for metazoan evolution. We also discuss inferences about the physiology and genecontent of the last common ancestor of animals and their closest unicellular relatives. We propose thatthe recorded Precambrian evolution of animals includes three intervals of advancement that begin withsponge-grade organisms, and that any preceding cryptic fauna would be no more complex than sponges. Themolecular data do not require that more complex animals appeared well before the recognized fossil record; nor,however, do they rule the possibility out, particularly if the interval of simpler metazoan ancestors lasted no morethan about 100 or 200 Myr. The geological record of abrupt changes in climate, biogeochemistry, andphytoplankton diversity can be taken to be the result of changes in the carbon cycle triggered by the appearanceand diversification of metazoans in an organic carbon-rich ocean, but as yet no compelling evidence exists forthis interpretation. By the end of this cryptic period, animals would already have possessed sophisticated systemsof cell–cell signalling, adhesion, apoptosis, and segregated germ cells, possibly with a rudimentary body planbased on anterior–posterior organization. The controls on the timing and tempo of the earliest steps in metazoanevolution are unknown, but it seems likely that oxygen was a key factor in later diversification and increasein body size. We consider several recent scenarios describing how oxygen increased near the end of thePrecambrian and propose that grazing and filter-feeding animals depleted a marine reservoir of suspendedorganic matter, releasing a microbial ‘clamp’ on atmospheric oxygen.

Received 18 February 2007; accepted 10 August 2007

Corresponding author: Eric J. Gaidos, Tel.: 1 808 956 7897; fax: 1 808 956 5512; e-mail: gaidos@hawaii.edu.

INTRODUCTION

The story of animal life began in the Precambrian geologicalera prior to 543 Ma, in a world that we would scarcelyrecognize as our own. Atmospheric carbon dioxide (CO

2

) wasmany times the modern value, but molecular oxygen (O

2

) wasmuch less abundant than today (Kasting & Catling, 2003).The land was devoid of plants, and the dominant visible forms

of life were mats of microorganisms forming laminated,organic–inorganic structures called stromatolites. We live ina very different world, one transformed by the activity ofmulticellular life; animals (Metazoa) as well as land plants. Theextant animal kingdom is composed of five major groups:the Porifera (comprising multiple clades of sponges), theenigmatic Placozoa (consisting of a single described species),Cnidaria (jellyfish, anemones, corals, hydrozoans, and seapens), Ctenophora (comb jellies), and the Bilateria (animalsdisplaying bilateral symmetry). The last three groups arecollectively known as the Eumetazoa. With the possible

T. Dubuc, M, Dunford, P. McAndrew, J. Padilla-Gamiño, B. Studer,and K. Weersing contributed equally to this work.

352

E. GAIDOS

et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

exception of acoels, which may be basal, the Bilateria aredivided into two groups based on major differences inembryonic development: the Protostomata (e.g. flatworms,roundworms, arthropods, molluscs, annelids), and theDeuterostomata (vertebrates, tunicates, lancelets, hemichordates,echinoderms). Most of the Protostomata have been furtherassigned to two ‘superphyla’, the Lophotrochozoa and theEcdysozoa (Halanych, 2004).

The transformation from the strictly microbial world toone shared by animals began in the Neoproterozoic Era(1000–543 Ma), and fossils of the Cambrian Period (543–490 Ma) include representatives of nearly all readily fossilizedmodern animal phyla (except bryozoans), as well as putativeexamples of now-extinct groups. Many authoritative reviewshave described and attempted to explain the geologically rapiddiversification of animal life recorded by Early and MiddleCambrian fossils such as those of the Chengjiang and BurgessShale deposits. For example, Marshall (2006) examinedpotential environmental, developmental, and ecologicalexplanations for the Cambrian ‘explosion’ and proposed thatthe diversification of animal body plans was the result of anincrease in the number of factors determining the fitnessof organisms as species began to interact with one another.Peterson

et al

. (2005) proposed that one particular interaction– macrophagy – was the most likely culprit. At the other endof the spectrum, Kirschvink & Raub (2003) speculatedthat the Cambrian explosion was triggered by a catastrophicgeophysical event – the inertial interchange of the Earth’srotation axis. Several syntheses, e.g. Valentine

et al

. (1999),Knoll & Carroll (1999), and Erwin (1999), have also addressedthe nature of the Cambrian radiation.

The oldest indisputable animal-like fossils are found at leastseveral tens of million years before the Precambrian–Cambrianboundary, but the time of emergence of the Metazoa as adistinct clade has not been established. Currently, all widelyaccepted metazoan fossils are confined to the latter half ofthe Ediacaran period. [The Ediacaran spans geological timebetween the terminus of a major glacial event 635 Ma and thePrecambrian–Cambrian boundary (Knoll

et al

., 2004).] Butone could argue

ad uniformitarian

that significant evolutionarychange requires lengthy geological time and thus recordedanimal history must have a substantial prelude. Indeed, somemolecular-based calculations of divergence times betweenextant metazoan groups yield ages that are much older thanthe oldest animal fossils (Wray

et al

., 1996). Although theestablished fossil record of animals is preceded by dramaticchanges in Neoproterozoic climate, atmospheric composition,the carbon cycle, and phytoplankton fossils, any relationship ofthese to the emergence of animal life remains conjectural.

In an insightful review, Budd & Jensen (2000) critiquedseveral scenarios that attempted to explain how Precambriananimals might have escaped fossilization, e.g. by being toosmall, too rare, or by occupying habitats where preservationdid not occur. They argued that important traits of Early

Cambrian body plans could not have evolved in organisms toosmall or too strictly planktonic to produce trace fossils, andconcluded that the appearance of such traits cannot be used toargue for an earlier, cryptic fauna. [Valentine (2002) madeessentially the same point.] Budd & Jensen (2000) explainedthe Cambrian ‘explosion’ by assigning many Cambrian fossilsto ‘stem’ groups that preceded the radiation of the ‘crown’groups (modern phyla), thus effectively minimizing the amountof evolutionary change required to produce them. ConwayMorris (2006) also critiqued the evidence for a prelude andargued that the ‘explosion’ was real and likely the inevitableresult of complex (but unspecified) ecological interactions.Early Cambrian fossils that can be unambiguously assigned tocrown groups (or their sister groups) could eventually testthese scenarios, but the fossil record is meager, e.g. Butterfield(1994) and Siveter

et al

. (2001).The early emergence of animal life remains a dynamic area

of study and new technologies such as whole-genome andexpressed sequence tag (EST) sequencing, multicollectorion-probe mass spectrometry, and X-ray tomography are usedto query the genomic and fossil record in ways that werehitherto impossible. Much of this research can be organizedas addressing four primary questions: Did the Metazoahave a lengthy Precambrian history? What were the climatic,biogeochemical, and ecological conditions during that interval,and how might the geological record be used to infer thepresence or absence of animals before trace or body fossils? Towhat extent can the physiological and genetic traits of extantanimal groups and their closest unicellular relatives be used toinfer the major evolutionary steps in an ancestral animal lineof descent? How might emerging metazoans have adapted toand altered their environment, i.e. the microbial biosphere,biogeochemical cycles, and, inevitably, other animals? Thisreview is organized accordingly.

THE RECORD OF PRECAMBRIAN ANIMAL LIFE

The fossil record

Recent palaeontological studies have refined the chronologyof Precambrian animal life, although they have not dramaticallychanged the overall picture. All uncontested animal fossils areno older than

c.

600 Ma and thus pre-date the Chengjiang(Early Cambrian) and Burgess Shale (Middle Cambrian)faunas by no more than 70–90 Myr. There are reports ofmuch older disks, burrows, and necklace-like forms that mighthave been left by animals (Hofmann

et al

., 1990; Seilacher

et al

., 1998; Rasmussen

et al

., 2002b; Fedonkin, 2003), butalternative interpretations such as microbial ‘mat’ structureshave been argued (Rai & Gautam, 1999; Seilacher

et al

., 1999;Conway Morris, 2002; Rasmussen

et al

., 2002a) and suchfinds await further substantiation, identification, and/or dating(Rasmussen

et al

., 2002c).

Emergence of animal life

353

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Among the oldest fossils of animal life are the remarkablephosphatized forms of the Doushantuo Formation in Chinathat have been traditionally assigned an age of around 580 Ma.The Doushantuo fossils include forms resembling benthic eggsand embryos – probably those of marine invertebrates (Chen

et al

., 2000; Xiao

et al

., 2000), rather than of algae (Xiao, 2002).Recent taphonomic experiments with sea urchin embryosdemonstrated that animal embryos could have been preservedlong enough for mineralization to occur (Raff

et al

., 2006).An alternative interpretation of these forms as giant vacuolatebacteria (Bailey

et al

., 2007) is not supported by the presenceof eukaryotic cyst-like organic vesicles around some fossils(Yin

et al

., 2007). More controversial are reports of cellularremains and body fossils, including forms that resemble epidermalcells, porocytes, amoebocytes, and sclerocytes of sponges, aswell as monaxid siliceous spicules that are characteristic ofcertain demosponges, e.g. (Li

et al

., 1998a); these may insteadbe fossil algae or products of diagenesis (Li

et al

., 1998b;Zhang

et al

., 1998). The purported spicules, in particular, lackaxial canals or excess silicon (Yin

et al

., 2001). Putative fossilsresembling the embryonic and adult forms of cnidarians havebeen described (Chen

et al

., 2002), and cross-sections of adultbilaterian bodies exhibiting a central gut bordered by pairedcoeloms have been reported (Chen

et al

., 2004b), but theseinterpretations are controversial, again because the forms mayrepresent diagenesis rather than biology (Bengtson & Budd,2004; Chen

et al

., 2004c). X-ray tomography of Doushantuofossils failed to show any of the epithelial organizationshared by all extant metazoans (Hagadorn

et al

., 2006),suggesting that these embryos represent earlier stem groups.

Recent radiometric dating and chemostratigraphy haveimproved estimates of the age and duration of the Doushantuoformation. The formation unconformably overlies a glacialdeposit that has been correlated to a 635 Ma unit in Namibia(Hoffmann

et al

., 2004). U-Pb dating of zircons from volcanicash beds that bracket the entire Doushantuo in one sectionindicates that it formed between 635 and 551 Ma (Condon

et al

., 2005). Zhang

et al

. (2005) dated an ash bed above thefossil-bearing uppermost phosphorites in another section to555 ± 6 Ma. Other dates are more equivocal. Whole-rock Pb–Pbdating of the stratigraphically older and younger parts ofthe upper (fossil-bearing) phosphorite unit yielded ages of599 ± 4 Ma (Barfod

et al

., 2002) and 576 ± 14 Ma (Chen

et al

., 2004a), respectively. Alternatively, Condon

et al

. (2005)correlated the subaerial exposure surface dividing the lowerunit (devoid of fossils) and upper Doushantuo unit with anepisode of glaciation 580 Ma, in conflict with the olderBarford

et al

. (2002) date, and placing the fossils between 550and 580 Ma. Kaufman (2005) notes that an unconformity andnegative excursion in the isotopic composition of inorganiccarbon at the

top

of the Doushantuo could instead be correlatedwith the same glaciation, making the fossils between 635 and580 Ma. The age difference between these different correlationsis thus as much as 85 Myr.

The oldest body fossils are imprints of diverse soft-bodiedorganisms (the Ediacaran fauna) whose taxonomic affinitiesremain controversial. Seilacher (1989) proposed that theEdiacaran biota represent a sister group (the ‘Vendazoa’) tothe Metazoa or at least a unique animal phylum (Buss &Seilacher, 1994) with synapomorphies such as a ‘quilted’structure. However, at least some of these organisms may havebeen members of extant metazoan phyla (Narbonne, 2005).The oldest reported fossils are of the frond-like

Charnia

thatappear immediately over a 575 ± 1 Ma ash bed in the Drookunit of the Avalon Formation in Newfoundland (Bowring

et al.

, 2003; Narbonne & Gehling, 2003) (Fig. 1).

Charnia

was originally interpreted as a macrophytic alga; that taxo-nomic assignment has been discounted based partly on itsdeep-water setting, and instead it has been identified as apennatulacean cnidarian (sea pen). Nevertheless, macrophyticalgae have been reported in waters as deep as 268 m (Littler

et al.

, 1985), and alternative assignments of

Charnia

(as spongeor colonial prokaryote) have been proposed (Steiner & Reitner,2001). Ediacaran taxonomy is controversial, and Brasier &Antcliffe (2004) recently questioned the basis (analogy) onwhich many fossils, especially

Charnia

, are classified. Theyunderscored the importance of distinguishing betweenmorphological diversity due to speciation or evolution withina group, and separation of body parts or a life cycle. With theexception of the Drook

Charnia

fossils, all Ediacaran fossilsare younger than about 565 Ma (Waggoner, 2003).

In a review of the history of the Ediacaran fauna, Narbonne(2005) made two salient points: First, before 560 Ma,ediacarans were immobile forms of limited diversity and aporiferan or cnidarian level of body plan (i.e. largely frond-likerangeomorphs such as

Charnia

). Investigators have arguedthat a few of these are stem-group ctenophores (Dzik, 2002;Shu

et al

., 2006). Second, mobile bilaterians formed part of ayounger (560–542 Ma), more diverse Ediacaran fauna thatpotentially contained representatives of all three bilateriansuperphyla (Fig. 1). For example,

Spriggina

was an arthropodor arthropod-like form (Ecdysozoa), annelid worm burrows,and probably the apparent mollusc

Kimberella

representedearly Lophotrochozoa. Less certain is the presence of earlyDeuterostoma:

Ernettia

, with chambered walls, may have beena chordate (Dzik, 1999), and

Tribrachidium

, with its distinctivethree-fold symmetry, has been seen as an echinoderm, but alsoas a cnidarian (Ivantsov & Fedonkin, 2002).

Disturbance or manipulation of sediments by animals,e.g. tracks, burrows, and changes in sediment fabric due tobioturbation, appear at least by 555 Ma (Martin

et al

., 2000)but no earlier than 560 Ma (Droser

et al

., 2002; Jensen, 2003;Jensen

et al

., 2005) (Fig. 1). The oldest unequivocal tracefossils are exclusively simple, unbranched, horizontal tracesthat formed close to the sediment–water interface, often justbeneath microbial mats (McNaughton & Narbonne, 1999;Jensen

et al

., 2005; Seilacher

et al

., 2005). Sulfidic conditionsbeneath the microbial mats that covered much of the seafloor

354

E. GAIDOS

et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

during that time appear to have restricted burrowing animalsto shallow depths within the sediment, and the absence ofdeep burrows until the end of Neoproterozoic time suggeststhat bilaterians had not yet evolved physiologies allowingthem to contend with these conditions, let alone benefit fromthem by feeding on sulfide-consuming bacteria as some taxado today (Bailey

et al

., 2006). Slightly more complex burrowsare known only from the latest Neoproterozoic (Jensen

et al

.,2000). One such trace fossil,

Treptichnus pedum

, definesthe base of the Cambrian (Landing, 1994) and coincideswith a significant increase in infaunal activity (Droser

et al

.,1999) (Fig. 1).

Skeletal biomineralization in animals first appeared in theEdiacaran but was restricted to silica and to hexactinellidsponges (Brasier

et al

., 1997), and possibly demosponges(Li

et al

., 1998a) (but see above). One of the earliest calcareousfossils was

Namapoikia

, consisting of somewhat irregulartubes packed closely together. It encrusted walls of fissures inmicrobial reefs of the Nama Group, Namibia,

c

. 549 Ma, andis considered likely to be a poriferan or cnidarian (Wood

et al

.,2002). The small shelly forms

Cloudina

and

Namacalathus

make a brief appearance immediately before the Precambrian–Cambrian boundary at 542 Ma (Amthor

et al

., 2003). Theshell of

Cloudina

is an irregular tube of stacked funnels.Despite its millimetre size, it is commonly found boredthrough, presumably by a smaller predator (Bengtson &Zhao, 1992; Hua

et al

., 2003), an observation supporting thehypothesis that biomineralization in animals was a defensive

adaptation.

Namacalathus

is a goblet-shaped form attachedto microbial reefs that has been interpreted to represent ametazoan with a cnidarian-like body plan (Grotzinger

et al

., 2000).In summary, a parsimonious interpretation of the fossil

record is that sessile, metazoan stem-group organismslacking epithelial organization and the ability to conspicuouslydisturb sediments appeared by ~600 Ma and perhaps asearly as 635 Ma. Members of some crown groups appearedby 580 Ma or shortly thereafter. Bilateria (or their stem-group representatives) appeared

c

. 560 Ma among the laterEdiacaran fauna and as evidenced by bioturbated sediments.Biomineralizing taxa appeared in the last few million years ofthe Precambrian and rose to prominence by the EarlyCambrian. Thus Precambrian animal life may have undergonethree successive intervals of evolutionary change, eachlasting about 15–20 Myr and involving (1) diversificationand epithelial organization (by 580 Ma); (2) bilateral symmetryand mobility (by 560 Ma); and (3) biomineralization andpredation (by 545 Ma). That the fossil record appears toresolve these advances suggests that earlier, unrecordedforms would have belonged to the first interval and beenlimited to organisms of a complexity no greater than non-biomineralizing sponges. Such a soft-bodied fauna wouldhave escaped fossilization and left no trace fossils. Loweroxygen levels may also have restricted the benthos to facieswhere sediment reworking made preservation less likely(Dornbos

et al

., 2006).

Fig. 1 A timeline of late Precambrian and earlyPhanerozoic history, styled after Narbonne (2005),with major geological (left-hand side) andbiological (right-hand side) events. The formerincludes major tectonic and climate events such asthe three major Neoproterozoic glacial intervals,and changes in the atmosphere. A schematic of theinorganic carbon isotope record is included. Shadedhorizontal bars represent the best available ageconstrains on the glacial intervals. Biological historyincludes highlights of the earliest animal fossilrecord, as well as changes in sediment texture dueto animal activity. The place of this interval in theentire geological column of Earth history isrepresented on the far right.

Emergence of animal life

355

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Molecular chronometry

In the method of the ‘molecular clock’ the rate of change inthe sequence of a gene is estimated using the divergence ofcopies in taxa whose first appearance in the fossil record isknown: the rate is then used to estimate the dates of deeperdivergences. Evolutionary genetics predicts clock-like evolutiononly for ‘neutral’ genes or parts of genes that are not subjectto selection, but this condition is usually not satisfied in casesof interest and clock-like behaviour is empirically demonstratedinstead. The first published divergence times between majormetazoan groups using this method pre-date the oldest animalfossils by as much as a factor of two (Runnegar, 1982a; Wray

et al

., 1996). Molecular divergence times should, in principle,be older than fossil dates (Bromham, 2003). Fossils give a

latest

time for the appearance of a new taxon – the Jaanussonor Sppil-Rongis Effect (Jaanusson, 1976) – and the actual agesof the calibration taxa and therefore the taxa of interest areolder. Although possibly important in the turnover of individualspecies on million-year timescales during extinctions, theeffect cannot explain discrepancies of tens to hundreds ofmillion years. Another explanation is the divergence of neutralgenes between isolated populations within a species beforeactual speciation and morphological divergence takes place.However, the amount of molecular variation involved is verysmall compared to the genetic divergence between the majoranimal groups.

This leaves two possibilities: the molecular clock calcula-tions are incorrect, or a cryptic fauna with a long evolutionaryhistory went unrecorded as fossils. Recent investigations haveattempted to improve divergence time calculations by includingmore taxa and more genes (Douzery

et al

., 2004; Peterson

et al

., 2004; Blair & Hedges, 2005b), employing morecalibration points (Peterson

et al

., 2004; Pisani

et al

., 2004),and using Bayesian statistics to account for uncertainties in, orthe use of lower limits for calibration times (Blair & Hedges,2005b). This attention has not produced broad agreement(Benton & Ayala, 2003); rather, it has revealed serious problemswith the molecular clock approach and its (mis)application,including the neglect of calibration and derivation errors(Graur & Martin, 2004), dependence on prior assumptions(Welch

et al

., 2005), and uncertainties in the underlyingevolutionary models (Roger & Hug, 2006).

Moreover, the assumption that the rate of molecularevolution is a constant independent of taxon or geologicaltime is now widely recognized as untenable (Bromham &Hendy, 2000; Bromham, 2003; Glenner

et al

., 2004; Ho

et al

., 2005; Pereira & Baker, 2006). A major criticism of earlywork was that calibrations based on slowly evolving vertebrategenes were used to date the divergences of their rapidly evolv-ing invertebrate counterparts, producing spuriously ancientages (Ayala

et al

., 1998). Peterson

et al

. (2004) and Peterson& Butterfield (2005) reported analyses that accounted forrelative rates and yielded dates for the protostome–deuterostome

divergence in much better agreement with the fossil record,although they arrived at this by rejecting the less congruentof two models. Blair & Hedges (2005a) showed that olderdivergence times are recovered using fossils ages as

minimum

rather than fixed calibration times. Although this is strictly true,they argue (circularly) that because divergence times of cladescan be much older than the oldest fossils, fossils themselvesprovide only a lower bound to calibration times; this dramaticrelaxation of the calibration dates naturally produces muchmore ancient divergence time for the uncalibrated clades.

Aris-Brosou & Yang (2003) went further, completely relaxingthe molecular clock assumption and replacing it with Bayesianprior models of rate change and speciation along a tree.They dated the protostome–deuterostome divergence to581 ± 112 Ma, in agreement with the fossil record. Douzery

et al

. (2004) used a similar analysis and reported an olderbut not significantly different time (761–642 Ma). Blair &Hedges (2005a) pointed out that these calculations alsoproduce other divergence times that violate the fossil record,and Welch

et al

. (2005) showed that the ages calculated withthis method may be unduly influenced by the assumptions ofthe model. While it seems certain that the molecular clockdoes not run true, there is either not enough information inour data to reliably account for its vagaries, or we do not yetknow how to extract such information (Bromham, 2006).

Taken together, these problems suggest that publishedconfidence intervals of molecular clock dates are greatlyunderestimated and that deep Precambrian divergence timesfor the major animal groups do not constitute a compellingargument for a lengthy pre-Ediacaran interval with a crypticanimal biota. Nor, however, can the molecular data or fossilrecord completely rule such an interval out, specifically oneinvolving organisms not much more complex than spongesand lasting no more than one or two hundred million years.What was the environment of these hypothetical earliestmetazoans, and might there be nonfossil clues to theirpresence in the geological record?

THE LATE PRECAMBRIAN WORLD BEFORE ANIMAL FOSSILS

Supercontinents and snowballs

Neoproterozoic climate was influenced by a lower solarluminosity, a phenomenon well founded in astrophysicaltheory (Gough, 1981), and compensatory greenhouse gases,whose past levels are much less certain. Most direct records orproxies of atmospheric composition are unavailable for thisera: However, Kaufman & Xiao (2003) interpreted carbonisotopes in Mesoproterozoic (

c

. 1.4 Ga) eukaryotic microfossilsas indicating CO

2

concentrations 10–200 times the presentatmospheric level (PAL). Such estimates are very uncertainbecause carbon isotope fractionation depends on many factorsbesides pCO

2

, including light level and the physiology and

356

E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

growth rate of the fractionating species (e.g. Laws et al.,1997). Elevated levels of methane could also have provided anenhanced greenhouse effect (Pavlov et al., 2003).

The late Precambrian Earth was not always warm, however.Harland (1964) first interpreted worldwide Neoproterozoicglacial deposits as evidence for intervals of global glaciation,but only recently have these been interpreted to reflectextreme ‘Snowball Earth’ events, in which an ice-albedofeedback of the climate system drove glaciation and pack iceformation to equatorial latitudes (Kirschvink, 1992; Hoffmanet al., 1998). This explanation has proved to be as controver-sial as it is dramatic, and the geological evidence is intenselydebated (e.g. Allen & Hoffman, 2005a; Jerolmack & Mohrig,2005; Allen & Hoffman, 2005b). Only a few accurate ages areavailable and more are clearly needed, but at least three majorintervals of glaciation have been identified: the Sturtian(c. 700 Ma) (Fanning & Link, 2004), the Marinoan (c. 635 Ma)(Hoffmann et al., 2004; Zhou et al., 2004), and the possiblyless severe Gaskiers glaciation (580 Ma) (Calver et al., 2004)(Fig. 1).

An important part of the controversy is the verificationof the glacial origin and global synchronicity of diamictiteunits. Eyles & Januszczak (2004) argued that many of thesedimentological features taken as evidence for Neoproterozoicglaciation (e.g. diamictites and lonestones) are not unambig-uously glaciogenic and could have formed in mass flows onunstable continental slopes during rifting. They propose that theunusually synchronous and global appearance of such sedimentswas a consequence of the accommodation space created bytectonic rifting during the break-up of the equatorial sup-ercontinent Rodinia, rather than climatic and glacioeustaticchange. In their view, the glacial sedimentary record is aconvolution of cold climate and tectonic extension, and theage and number of actual glacial episodes may not be amenableto correlative dating (Kennedy et al., 1998). The break-up ofRodinia was previously thought to have occurred at around750 Ma; however, new palaeomagnetic data suggest that itsdisintegration began 50–100 Myr earlier (Fig. 1) and wascomplete by 750 Ma (Torsvik, 2003). This is too early toexplain the formation of glacial sedimentary units by thecreation of accommodation space (Eyles & Januszczak, 2004).The earlier date also challenges several proposed triggermechanisms for major Neoproterozoic glaciations based ontectonics associated with the rifting of the supercontinent(Hoffman, 1999; Ridgwell et al., 2003; Donnadieu et al.,2004). Nonetheless, an earlier demise of Rodinia may haveinitiated a cooler period in which the Earth was more sus-ceptible to glaciation.

Also controversial is the latitudinal extent of glaciationand whether oceans were completely ice-covered (the ‘hard’snowball scenario) or thin ice or open seas persisted at theequator (the ‘slushball’ scenario) (Warren et al., 2002; Good-man & Pierrehumbert, 2003; Pollard & Kasting, 2005).Negative feedbacks in the climate models that include the

global hydrological cycle and the effects of clouds resistextreme cooling such that the most severe snowball stateis difficult to enter (Hyde et al., 2000) or escape from (Pierre-humbert, 2004). Two different, apparently contradictory linesof evidence concerning ice cover have been recently reported.The first involves the presence of biomarkers, specifically2-α-methylhopanes (from cyanobacteria) and alkylated2,3,6-trimethylbenzenes (derived from the isorenieratenepigments of green sulfur bacteria), in glacial deposits in a corefrom South America (Olcott et al., 2005). The former constituteevidence that photosynthesis was taking place in the glacialoceans, while the latter indicates the presence of sulfide in thephotic zone. The presence of phototroph biomarkers in theglacial units is not entirely surprising, given that the canonicalsnowball scenario posits an interval of intense glaciation (andtillite deposition) before the climactic snowball event in whichthe oceans freeze over, the hydrological cycle shuts down, andmarine sediments stop accumulating. A ‘hard’ snowball intervalitself is represented by a hiatus in the sedimentary record andis not amenable to traditional analysis.

In another study, elevated concentrations of iridium (Ir) ator near the base of carbonates overlying Sturtian and Marinoanglacial deposits in Africa are interpreted as the result ofaccumulation of extraterrestrial matter over several millions ofyears of snowball glaciation (Bodiselitsch et al., 2005). This Iris supposed to have fallen in interplanetary dust particles ontothe pack ice and to have subsequently been released into thewater column when the ice melted and marine depositionresumed. There are, of course, terrestrial sources of platinum-group elements such as Ir, and concentration spikes in marinesediments can be produced by changes in sedimentary redoxconditions (Colodner et al., 1992). In principle, a terrestrialorigin has been excluded by the behaviour of multiple platinum-group elements (Bodiselitsch et al., 2005), but measurementsof other indicators of an extraterrestrial source, e.g. osmiumisotopes (Peucker-Ehrenbrink & Ravizza, 2000), are clearlyneeded. Ir spikes could also be produced by one or more largeimpacts; one such event created the post-Marinoan Acramanimpact structure in Australia (Grey et al., 2003).

The origin of ‘cap’ carbonates – dolostones that directlyoverlay glacial units and formed in a marine transgression – isundoubtedly one key to understanding the nature of Neopro-terozoic glaciations (Shields, 2005). They were originallyexplained by the postglacial overturn of anoxic oceans andinjection of deep-water CO2 into the surface ocean (Grotzinger& Knoll, 1995). In the classical snowball scenario, they areinstead the product of intense weathering during the hothouseconditions that terminated the glacial period, and the resultingabiotic precipitation of accumulated atmospheric CO2 ascarbonates (Hoffman et al., 1998). Shields (2005) argued thatthis and other models do not explain the global uniformity andcomplexity of cap carbonate sequences, and proposed insteadthat the carbonates were precipitated by algal blooms in amassive, transient freshwater layer produced by rapid glacial

Emergence of animal life 357

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

melting. A less catastrophic explanation of these enigmaticunits is that they are the consequence of increased alkalinity ina glacial ocean with greatly reduced neritic (shelf) area at atime when calcifying organisms were not present to driveprecipitation of carbonates in the pelagic zone (Ridgwell et al.,2003). One means to distinguish between these hypotheses –one in which carbonates were precipitated under high temper-atures and the others in which they were precipitated undernear-freezing conditions – might be carbonate palaeother-mometry, e.g. a search for ikaite pseudomorphs (glendonites)such as those that have been found in carbonates between theglacial intervals (James et al., 2005).

Another explanation for the cap carbonates is that theircarbon originated as biogenic, isotopically light methane inpermafrost that was released and oxidized to CO2 by floodingduring a postglacial marine transgression. Kennedy et al. (2001)noted that the amount of carbon needed to produce thecap carbonates is consistent with that required to producecontemporaneous, large negative carbon isotope excursions,provided that all the carbon is from isotopically light biogenicmethane (δ13C ≈ –60‰ relative to the Pee Dee belemnite, seediscussion of the carbon isotope record in the next section).They argued that a colder climate, sea-level drawdown, andadditional shelf area in the post-Rodinia world meant that theNeoproterozoic methane hydrate reservoir was at least aslarge, if not larger than that on the modern Arctic shelf.However, sufficient organic matter may not have accumulatedin soils in the absence of land plants, or in marine sedimentsexposed during the preceding glacial regression. Also, thisscenario does not explain why the carbonate units invariablycap the glacial units, as glacial activity could continue even ascontinental ice sheets retreated and sea level rose. Hoffman &Schrag (2002) also claimed that sulfate might have limitedanaerobic oxidation of the methane, although the amount ofmethane oxidized (8 × 1017 mol) is dwarfed by the total sulfurbudget of the oceans (4 × 1019 mol).

The carbon cycle

Late Neoproterozoic carbonate rocks, including the ‘cap’carbonates, also contain the largest known fluctuations(11–15‰) of carbon isotopic composition (δ13C) in Earthhistory (Fig. 1). These follow a relatively featureless billion-year period in the carbon isotope record, described by RogerBuick as the ‘boring billion’. Isotopic excursions associatedwith the Sturtian and Marinoan glaciation can be explained bya ‘hard’ snowball scenario in which primary production is shutoff by pack ice (Hoffman et al., 1998). However, new modelssuggest that ice may have been thin or absent at low latitudes(Warren et al., 2002) and that the glacial oceans would havebeen nutrient rich, enhancing productivity. The snowballscenario also does not explain those excursions not associatedwith glaciations, i.e. at 550 Ma and at the Precambrian–Cambrian boundary itself (Aharon, 2005; Zhang et al., 2005).

Ocean overturn and mixing are another possible explanationof the excursions in inorganic δ13C (Aharon, 2005). However,this requires a large isotopic gradient with depth, and theresidence time of inorganic carbon in the shallow ocean (thesource of the carbonates) is years (Kump, 1991) and far tooshort to explain anomalies that are estimated from subsidencerates to have persisted for 0.1–1 Myr. Regardless, the effect onthe ratio of organic matter to inorganic burial rate, and hencethe inorganic carbon isotopic composition would be small(Halverson et al., 2002).

Injection of isotopically light carbon into the ocean–atmos-phere system could explain the anomalies (Bartley et al., 1998).Halverson et al. (2002) proposed a prolonged release ofbiogenic light methane from clathrate hydrates to the atmospherewhere it was oxidized to CO2. Walter et al. (2000) linkdestabilization of marine clathrates to warming at the end ofglaciations: One problem with this scenario is that warmingis accompanied by a eustatic rise that stabilizes clathrates(a 100-m rise offsets a 3 °C increase), and the break-up ofRodinia would have exacerbated the problem by furtherelevating sea level. As discussed above, Kennedy et al. (2001)instead invoked flooding and thawing of a terrestrial perma-frost reservoir which they argue could have been more than ahundred times larger than that found at present.

Organic matter is also isotopically light (δ13C ≈ –25‰):Rothman et al. (2003) showed that the Neoproterozoiccarbon isotopic record cannot be explained by carbon cyclemodels that have been successfully applied to Phanerozoictime. They proposed instead that the Neoproterozoic oceanscontained 2–3 orders of magnitude more reactive (dissolvedor suspended particulate) organic matter than present, andthat negative carbon isotopic anomalies could be explained bybiotic remineralization of this reservoir and exchange withinorganic carbon in the ocean and atmosphere. Elevated levelof reactive organic matter were permitted by ocean stratificationand increased organic matter suspension, lower oxygen, andthe absence of filter-feeding and/or predation by animals.

In fact, the billion years preceding the late Neoproterozoicisotope events might include important but undetectablechanges in the carbon cycle and not be boring after all. Bartley& Kah (2004) argued that a much larger dissolved inorganiccarbon (DIC) reservoir in the ocean–atmosphere systemwould have masked any perturbations to the carbon cycle.Such a reservoir would have existed if ocean pH was constant(Grotzinger & Kasting, 1993) but atmospheric CO2 washigher because of a silicate weathering feedback acting undera fainter Sun (Walker et al., 1981). A large marine DIC poolwould also have buffered changes in the partial pressure ofCO2 (pCO2) and its greenhouse effect, explaining the absenceof glaciations before the Neoproterozoic. As pCO2 and thesize of the marine DIC reservoir declined, the sensitivity of thesystem to fluctuations in the rate of remineralization ofdissolved organic carbon (DOC) increased, particularly if DOCfar exceeded the modern reservoir (Rothman et al., 2003).

358 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Oxygen and ocean chemistry

The availability of molecular oxygen is widely considered tohave been the pacemaker of geochemical change and biologicalevolution during Precambrian time. The near-consensus ofpalaeo-Earth scientists is that free O2 was virtually absent fromthe early Precambrian atmosphere but appeared in small butgeochemically significant concentrations (at least 0.1% PAL)by about 2.4 Ga (Canfield, 2005). A geochemical modelbased on sedimentary rock abundances and isotopes of C andS predicts that the atmospheric O2 did not vary by more thana factor of two during the Phanerozoic (543 Ma to present)(Berner et al., 2003). pO2 during the intervening (Proterozoic)time is therefore most pertinent to the rise of animals but isalso poorly constrained. Euxinic conditions in the oceanlimit Proterozoic O2 to 3% PAL, but this estimate dependson the assumed phosphorus concentration and rate ofocean turnover (Kasting, 1987). Enhanced rates of organicmatter burial inferred from carbon isotope analysis suggestpO2 rose at 800–580 Ma (Des Marais et al., 1992). Thistiming (Fig. 1) is consistent with an increase to perhaps 5–18%PAL inferred from a depletion of 34S in biogenic marinesediments and a molecular clock estimate of the proliferationof nonphotosynthetic sulfide-oxidizing bacteria (Canfield& Teske, 1996). However, given the serious problems withmolecular chronometry, the second of those argumentsshould be set aside.

Changes in ocean redox state, thought to reflect atmosphericoxygen, are recorded in the iron content of Precambriansedimentary rocks. Banded iron formations (BIFs) are cos-mopolitan Precambrian sedimentary units consisting ofalternating layers of iron oxides (magnetite and haematite)and chert. The trace and rare earth element patterns inProterozoic BIFs are consistent with a deep-sea hydrothermalsource (Klein, 2005). In the anoxic deep oceans of the earlyPrecambrian (prior to 1.8 Ga), Fe existed in its soluble ferrousform and was readily transported to shallow continentalshelves, where it reacted with oxygen (produced by marinecyanobacteria) and precipitated as ferric iron. [An alternativeexplanation involves anoxygenic photosynthesis by bacteriausing ferrous iron as an electron donor (Kappler et al., 2005).]Banded iron formations are thus an indication of the pastchemical state of the deep oceans.

The disappearance of BIFs after 1.8 Ga (until their briefre-appearance in association with Neoproterozoic glaciations)was originally attributed to oxygenation of the deep sea andimmobilization of iron (Beukes & Klein, 1992). Canfield(1998) proposed an alternative explanation in which the deepProterozoic ocean remained anoxic and became sulfidic after1.8 Ga: Increased subaerial weathering of sedimentary pyrite(FeS2) under an O2-containing atmosphere led to elevatedmarine sulfate, production of sulfide, titration of iron from theoceans as pyrite, and suppression of BIF formation (Fig. 2).Evidence for the ‘Canfield Ocean’ includes elevated sedimentary

pyrite at 1.84 Ga (Poulton et al., 2004) and an increase insulfur isotopic fractionation between sedimentary pyrite andgypsum beginning 2.2–2.3 Ga – attributed to the action ofsulfate-reducing bacteria on sulfate concentration above1 mM (Canfield, 1998). Still, high inferred rates of S isotopechange in Proterozoic sediments suggest that sulfate levels didnot exceed a few mM and were still an order of magnitudebelow the modern value (Shen et al., 2003; Kah et al., 2004).Biomarker data also support Canfield’s scenario: Logan et al.(1995) argued that the unusually complete biodegradationof algal compounds during the Proterozoic was in part due tothe action of sulfate-reducing bacteria. Early Proterozoicsedimentary rocks also contain carotenoids derived fromphototrophic purple sulfur bacteria and green sulfur bacteria,indicative of euxinic conditions (Brocks et al., 2005) (Fig. 2).

The reappearance, after a billion-plus year hiatus, of iron-rich sediments with Neoproterozoic glacial deposits andextreme carbon isotope excursions can be interpreted either asa return to total global marine anoxia due to the isolation ofthe oceans from the atmosphere (Kirschvink, 1992), or, alter-natively, a decrease in the weathering rates of sedimentarypyrite and flux of sulfur as sulfate into the oceans as a conse-quence of a colder climate (Canfield & Raiswell, 1999). Theresidence time of sulfur is about 3 Myr in the modern ocean(and perhaps shorter in the Proterozoic), and thus marinesulfur concentration could have been responsive on thattimescale.

The fossil record of phytoplankton and its implications

In the absence of obvious abiotic triggers, it is tempting torelate Neoproterozoic perturbations in the carbon cycle (andperhaps climate) to events in biological evolution. Logan et al.(1995) proposed that the rapid removal of organic matterfrom the water column in the form of faecal pellets, beginningin Cambrian time, resulted in the improved preservation ofalgal lipids. In the Precambrian, smaller particles sank moreslowly and organic matter would have been rapidly degradedin the water column. Logan et al. (1995) compared hydrocarbons(the diagenetic products of organic matter from planktonicmicroorganisms) in sedimentary rocks of Neoproterozoicage with those from the Phanerozoic. They found thathydrocarbons derived from heterotrophic bacteria dominatein Neoproterozoic sedimentary rocks but are largely replacedby those derived from photosynthetic algae in younger rocks.Autotrophic metabolism produces C14-C18 n-alkyl carbonchains that are 13C-depleted compared to co-occurringisoprenoids pristane and phytane, both of which arephotosynthetic degradation products. Heterotrophic metabolismproduces the opposite isotopic shift (Hayes, 2001). A distinct13C depletion in n-alkyl compounds is observed through thePrecambrian–Cambrian boundary, indicating a shift fromefficient carbon cycling in a heterotrophic microbial loop tomovement of that carbon to higher trophic levels, e.g. grazers

Emergence of animal life 359

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

and predators, and its eventually loss to the seafloor as faecalpellets (Logan et al., 1997). An opposite shift accompanies themass extinction of animals at the Permo–Triassic boundary(Grice et al., 2005). This is consistent with the proposal ofRothman et al. (2003) that a large marine reservoir of reactiveorganic carbon disappeared in the late Neoproterozoic,perhaps as a consequence of the evolution of filter-feeding,predation, and/or the development of animals with gutscapable of faecal pellet production.

Clues to the carbon cycle–animal connection might lie inthe Neoproterozoic fossil record of eukaryotic phytoplankton,which consists largely of acritarchs, organic-walled planktonicmicrofossils of problematic and probably polyphyletictaxonomic affinity. Their record reflects global environmentalchanges that presumably influenced the early diversification ofanimal life as well. Furthermore, phytoplankton and theanimals that consumed them must have influenced eachothers’ evolution and extinction (Butterfield, 2007). Prior tothe Marinoan glaciation, individual acritarch species typicallypersisted for hundreds of millions of years. At around 600 Ma,species diversified in size and morphology, and longevitiesdeclined to 15–50 Myr (Knoll, 1994; Grey et al., 2003;Huntley et al., 2006). The immense longevity of the pre-Marinoan acritarch species may have resulted from saturationof the photic zone by populations of cyanobacteria and

phytoplankton in the absence of effective predation andherbivory. Under such conditions, nutrients would havelimited population size and intense competition would haveled to severe niche partitioning, leaving few opportunitiesfor new species (Stanley, 1976). The emergence of metazoanzooplanktors sometime after the Marinoan glaciation butbefore the earliest record of the Ediacaran fauna may haverelaxed nutrient limitation, allowed new species to invade, andapplied new selection pressures, producing the observed shiftin morphotype diversity and longevity (Butterfield, 1997;Peterson & Butterfield, 2005). Indeed, one acritarch morpho-type corresponds to the embryo-containing vesicles in theDoushantuo, suggesting that metazoans may have existed by632 Ma (Yin et al., 2007).

Acritarchs declined dramatically in both diversity andmorphologic variety at about the time of appearance of theEdiacaran fauna (c. 575 Ma), and they failed to recover untilCambrian time (Knoll, 1994). The earliest members ofthe Ediacaran fauna included sessile suspension feeders (i.e.rangeomorphs) (Clapham et al., 2003) and thus a causalconnection between these biotic events, one entailing a groupof producers and the other a group of primary consumers, is aplausible, albeit hypothetical scenario. It is also possible thatenvironmental change somehow simultaneously favoured theevolutionary radiation of metazoans, while being detrimental to

Fig. 2 Important biogeochemical processes in a hypothetical Neoproterozoic ocean and in terrestrial weathering: in this schematic, arrows identify chemical flowsconnecting these processes. The deep ocean is assumed to be anoxic and sulfidic; euxinic presence of sulfide in the photic zone is exploited by anoxygenicphototrophs. The oceans may have been an important source of atmospheric H2S and the greenhouse gases CH4 and N2O. POM = particulate organic matter;DOM = dissolved organic matter.

360 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

acritarchs. Grey et al. (2003) correlated the acritarch decline withthe Acraman impact in Australia, but it is not obvious why suchan impact would not have also affected adversely animal life.

The record of Proterozoic carbonaceous compressionsinterpreted to be macroscopic algae also suggests stepwisechanges in diversity and morphology contemporaneous tothose in the acritarch record (Xiao & Dong, 2006). Becauseof the difficulty of distinguishing algal from microbial car-bonaceous compressions and of determining whether suchchanges were driven by animal herbivory or by some otherfactor such as nutrient limitation, caution should be exercisedin interpreting the fossil macroalgae record. Nevertheless,the geological record of the Neoproterozoic does containunambiguous indications of dramatic changes in the carboncycle, some temporally connected to glaciations, and it alsoprovides intriguing evidence that a reorganization of themarine food web occurred. If ancestral animals were somehowinvolved in these changes, what may have they been like? Inthe absence of a fossil record we must turn to the physiologies,life cycles, and genomes of living animals for clues.

THE LIVING RECORD OF EARLIEST ANIMAL EVOLUTION

Extant organisms are a rich source of information about theorigins and early evolution of animals, one that is complementaryto the fossil record. This information can be used in threeways: First, extant single-celled organisms can be studied asanalogs to the eukaryotic ancestors of metazoans. Second,comparisons of metazoans with their closest unicellularrelatives can be used to infer genetic traits of a hypothetical‘protometazoan’, the last common ancestor of both groupsand the starting point of uniquely metazoan evolution. Finally,comparisons among all extant animal groups can be used toinfer the properties of the ‘urmetazoan’ and eumetazoanancestors of all extant metazoans and of all nonspongemetazoans, respectively.

A unicellular beginning

The oldest eukaryotic fossils with a taxonomic affinity of anycertainty are of the red alga Bangiomorpha from 1.2 Ga rocks(Butterfield, 2000; Javaux et al., 2001), although some oldermacroscopic fossils may be of eukaryotes (Walter et al., 1990;Kumar, 1995). Steranes, the saturated degradation productsof sterols, have been found in 2.7 Ga shales (Brocks et al.,1999). These are widely thought to be of eukaryotic origin,although the possibility that they are from cyanobacteria hasbeen raised (Volkman, 2005) (but see Summons et al., 2006).Thus, unicellular eukaryotes were present at least half a billionand perhaps as long as 2 billion years before the emergence ofmetazoans. The Neoproterozoic fossil record of eukaryotictaxa thought to be closely related to the Metazoa is poor(Fig. 1): Fossil testate amoebas have been found in the c. 750 Ma

Chuar Group in Grand Canyon (Porter & Knoll, 2000; Porter& Knoll, 2003). The oldest established fossils of fungi (themajor sister group to the Metazoa) are of Ordovician age(Redecker et al., 2000), although possible Neoproterozoicexamples have been reported (Butterfield, 2005; Yuanet al., 2005).

Phylogenetic studies place the origin of the Metazoa amonga group of unicellular protists, some of which have coloniallifestyles (see following section). However, multicellularity orat least colonial behaviour has evolved in multiple eukaryoticgroups and in some bacteria, presumably due to the advantagesuch lifestyles provide for feeding, dispersion and protectionfrom predation (Bonner, 2000; Kaiser, 2001; King, 2004;Kirk, 2005). One idea is that multicellularity emerged byadaptive resolution of conflict between competing cells, i.e.cooperation enforced by sanctions against ‘cheaters’ (Buss,1999). Colonial cells produce signals, i.e. small molecules thatare mediated by the environment and processed by a signaltransduction system involving transmembrane proteins such ashistidine and tyrosine kinases. They also communicate bydirect interactions between transmembrane proteins and bygap junctions that directly connect the cytoplasm of adjacentcells (Kaiser, 2001). Hazkani-Covo et al. (2004) comparedthe complete genomes of the unicellular yeast Saccharomycescerevesia to those of the bilaterians Drosophila melanogasterand Caenorhabditis elegans. In the latter two organisms theyfound a proliferation of genes encoding proteins targeted tothe cell membrane and exterior. They interpreted this as theresult of selection for more sophisticated cell–cell communicationduring the emergence of animals.

Cellular differentiation is also widespread in protozoans andalgae. It has been studied in detail in the colonial alga Volvoxcarteri, which displays germ-soma division of labour (Kirk,2005), the social amoeba Dictyostelium discoideum, whichforms fruiting bodies and mobile ‘slugs’ (Eichinger et al.,2005), and the unicellular protozoan Naegleria gruberi, whichcan repeatedly alternate between amoeboid and free-swimmingflagellate forms (Fulton, 1993). It seems likely that theunicellular lineage leading to the Metazoa was already capableof cell–cell communication and cellular differentiation. Müller(2003) proposed that these were two of many steps in theevolution of an autonomous, unicellular eukaryote into a primitivemetazoan ancestor, an integrated colony or individualcomposed of cooperating, differentiated cells with theorganizational grade of a simple sponge.

The protometazoan ancestor: from many, one

Molecular phylogenetic analyses show that the Choanoflagellates,a monophyletic group of protists of which some possesscolonial lifestyles, are the closest sister group to the Metazoa,with which they form a clade to the exclusion of the fungi(Medina et al., 2001). Choanoflagellates also have amorphological resemblance to the flagellated choanocytes of

Emergence of animal life 361

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

sponges. Maldonado (2004) argued that choanoflagellates arenot originally unicellular but are instead highly derivedsponges that have reverted to a single-celled existence. Onetest of this hypothesis is the relative size of the mitochondrialgenomes of choanoflagellates, sponges, and other metazoans,which are thought to be under strong selective pressurefor downsizing: If choanoflagellates are extremely derivedsponges, they should have mitochondrial genomes the samesize or smaller than those of sponges. Instead, the mitochondrialgenomes of two sponge species are intermediate in sizebetween those of choanoflagellates and other metazoans(Lavrov et al., 2005), consistent with sponges evolving from achoanoflagellate-like ancestor. Two choanoflagellate species,Monosiga brevicollis and a Proterospongia-like species, expresshomologs to animal genes for cell–cell signalling and adhesionsuch as tyrosine kinases and G protein-coupled receptors. Atleast some animal cell adhesion factors, i.e. cadherins, and C-type(calcium-dependent) lectins are present in choanoflagellatesand thus predate metazoans (King et al., 2003). Protein–proteininteraction domains also previously thought to be unique tometazoans have been found in choanoflagellates (King,2004). It would appear that many of the biological buildingblocks for a multicellular existence were already present in thelast common ancestor of the choanoflagellates and metazoans.

A second group occupying a pivotal phylogenetic locationwith the choanoflagellates near the animal–fungal node is theMesomycetozoea, a group of parasitic or saprophytic microor-ganisms (Mendoza et al., 2002). Some members of this grouphave characteristics of both protists and fungi, and complexlifecycles driven by interactions with their hosts. However, it isnot known if the phylogenetic placement of mesomycetozoeanshas been somehow affected by their parasitic lifestyle, as appearsto be the case for some fast-evolving parasites such as Giardialamblia. Metazoans, choanoflagellates, and mesomycetozoansare sometimes collectively termed the Holozoa, but it is not yetestablished that this grouping represents a monophyletic clade.

Reconstructing the urmetazoan ancestor

Reconstruction of evolutionary events within the Metazoa iscontingent on an accurate phylogeny. The use of protein andeventually gene sequences and other genetic markers inquantitative approaches (Zuckerkandl & Pauling, 1965) haspermitted a systematic, although not necessarily complete orincontrovertible view of animal phylogeny (Adoutte et al.,2000; Halanych, 2004). Recent technical advances ingenomics and computational biology have greatly expandedthe phylogenetic repertoire of genetic markers, taxa, andanalytical tools. The last includes the improvement ofcomputational economy in maximum likelihood techniques(Schmidt et al., 2002) and the development of methods basedon Bayesian analysis (Huelsenbeck et al., 2001).

The different methods yield variation in the placements ofphyla within the overall tree of animal life, but have consistently

resolved several major groupings; the Bilateria, Cnidaria,Ctenophora, multiple clades of poriferans, and Placozoa(Fig. 3). The phylogenetic relationships between these fiveclades are still not resolved (Zrzavy et al., 1998; Wallberget al., 2004), although most analyses place poriferans at thebase of the animal tree. Recent work indicated that spongesare paraphyletic and that the Eumetazoa are sister to thecalcareous sponges, to the exclusion of the hexactinellidsponges and demosponges (Borchiellini et al., 2001; Peterson& Butterfield, 2005). The branching order of cnidarians,bilaterians and ctenophores is unresolved, although moleculardata on the last group are only now becoming available. Theplacement of placozoans is likewise problematic: One analysisplaced them as sister to the Bilateria (Wallberg et al., 2004),while a recent analysis of the placozoan mitochondrial genomeconcluded that placozoans are basal to sponges – and all othermetazoans (Dellaporta et al., 2006). The resolution of theserelationships will be crucial to constructing an explanatoryhistory of evolution of the Metazoa.

The lack of resolution among basal groups of metazoansraises the question of whether this polytomy has evolutionarysignificance or is simply a reflection of the limits of our dataand/or methods. Using independent simulations, Levintonet al. (2004) and Rokas et al. (2005) found that radiationswhich occurred in a relatively short period of time in thedistant past will likely result in irresolvable molecular phylogenies.It is possible that our inability to resolve the precise relationshipsbetween several groups of phyla indicate that they divergedrapidly. It is also possible, however, that these unresolvedpolytomies are an artefact of the selection of the genes and taxaused in the studies (Jermiin et al., 2005).

Setting aside the question of the position of the Placozoa,phylogenetic reconstructions taken at face value suggest thateumetazoans are derived and that the urmetazoan ancestorwould have many traits found in poriferans. Some bilateriancell signalling and adhesion genes are expressed in the spongeOscarella carmela (Nichols et al., 2006). Larroux et al. (2006)reported that some developmental genes found in bilateriansare also expressed during development in the demospongeReniera, even though sponges are thought to possess a primitivegrade of developmental complexity (Brusca & Brusca, 2003).Such conserved gene functionality suggests that the urmetazoanancestor had the potential for specification of multiple cell types,formation of fixed body axes, and simple tissue development(Larroux et al., 2006). The urmetazoan ancestor would havehad mechanisms to control growth and eliminate unwantedcells in those tissues (Müller, 2003). Multicellular organismspossess a variety of mechanisms that regulate their size or thatof their constitutive tissues (Gomer, 2001), but animalshave widely adopted apoptosis (programmed cell death) asan integral part of tissue development and size regulation.Apoptosis involves the synthesis of enzymes such as caspases,a family of cysteine proteases that break down cells. Caspasesare present in both sponges and cnidarians (Wiens et al.,

362 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

2003a; Wiens & Müller, 2006) but their origin is unclear. Notrue caspases have been yet found in eukaryotes outside Metazoaalthough caspase-like activity has been documented in plants,fungi, protists, and some bacteria, and genes with significantsequence similarity to caspases have been discovered in thesegroups (Boyce et al., 2004). [The animal form of caspase mayin fact be the result of horizontal transfer from bacteria(Koonin & Aravind 2002)]. The sponge-like urmetazoanancestor would also have had segregated germ cells. Animalssegregate germ cells by two mechanisms; epigenesis (signalinduction) and preformation by inherited determinants.The former is used to the exclusion of the latter in basal groupssuch as sponges and cnidarians and would appear to beancestral (Extavour & Akam, 2003). In these organisms,pluripotent cells can produce both germ and somatic stemcells, allowing for regeneration as well as both sexual andasexual reproduction.

From one, many

Most of metazoan diversity is based on developmental pathwaysin which stem cells differentiate into multiple somatic cell

types in specific spatial patterns. This often takes place duringembryogenesis from zygote to adult, although sponges andjellyfish and some bilaterians retain such totipotent cells inthe adult form and can reproduce by budding. Since thepioneering 19th century work of T. H. Huxley and E. Haeckel,and later research by Hyman (1940), differences in thephysiology of extant organisms as well as the evolution of bodyplans have been cast in terms of differences in developmentand of the evolution of regulatory genes involved in development(Goodman & Coughlin, 2000).

The introduction of genomic and proteomic techniques hasrevitalized the study of developmental evolution. A typicalapproach is to compare the pattern of spatial expression ofdevelopmental genes in two evolutionarily distant organismsand to infer the function of the gene in their last commonancestor (Martindale et al., 2004; Martinelli & Spring, 2004;Larroux et al., 2006). However, conservation of spatialexpression pattern does not necessarily imply conservation offunction (Nielsen & Martinez, 2003; Svensson, 2004). Whileit is relatively easy to verify gene function in model organismssuch as Drosophila melanogaster which have well-developedgenetic tools, such finesse is not available for many invertebrates.

Fig. 3 Schematic phylogeny of the major extant groups of the Metazoa and their closest relatives, the choanoflagellates and mesomycetozoan protists. The threegroups are collectively termed the Holozoa. The Fungi is the sister clade to the Holozoa, to the exclusion of plants and all other unicellular eukaryotes. Thephylogenetic placement of Placozoa is controversial, with different analyses placing it basal to all other Metazoa, basal to the eumetazoan ancestor, or as a sistergroup to Bilateria. The evolutionary relationships between Bilateria, Cnidaria, and Ctenophora are also not established, and the paraphyletic nature of sponges isstill tentative.

Emergence of animal life 363

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

In such cases, gene silencing by small interfering RNA(siRNA) (Hannon, 2003) could prove to be a useful probe ofdevelopmental gene function. Interpretation of experimentsalso depends on the phylogenetic placement of the organismsof choice, something that is controversial for many basalmetazoan taxa. Martindale (2005) notes that much of thework has been done on highly derived organisms that mayhave lost many primitive traits relative to basal taxa.

A common conclusion of studies so far is that developmentalgenes and pathways (or their homologs) are highly conservedacross all major phyla, including basal nonbilaterians. Thewidespread conservation of both embryogenesis pathways anddevelopmental regulatory genes also suggests that the core ofdevelopmental gene regulatory networks was ‘frozen’ earlyin animal evolution (Davidson & Erwin, 2006) and thatevolutionary events responsible for different body plans andphyla happen by comparatively modest rearrangements orduplications of existing genetic machinery (Gilbert, 2003).

Members of the Hox gene family of homeodomain proteinsplay a key role in pattern formation during bilaterian embry-ogenesis, and the diversification of triploblastic (bilaterian)animal body plans has been interpreted in terms of the changingnumber, specificity, and expression of clusters of these genes(Gellon & McGinnis, 1998; Peterson & Davidson, 2000; Garcia-Fernández, 2005a, b). The picture has been complicated by thediscovery of the ParaHox sister group (Brooke et al., 1998) andthe finding that cnidarians, diploblasts with only rudimentary(or vestigial) symmetry, possess both Hox and ParaHox genes(Finnerty & Martindale, 1999). Hox/ParaHox genes appearto be unique to the metazoans, but non-Hox homeoproteinsare found in plants and fungi where they play a role in matingtype regulation. The presence of Hox and ParaHox genes inboth the Bilateria and the Cnidaria indicates that an ancestral‘ProtoHox’ gene cluster diverged in an even more distantmetazoan ancestor (Chourrout et al., 2006; Ryan et al.,2007). Searches for ProtoHox analogs in ctenophores andplacozoans have uncovered promising candidates (Finnertyet al., 1996; Jakob et al., 2004); there is at least one home-oprotein in sponges (Perovic et al., 2003), but so far no Hoxgenes have been found. If further genome sequencing inadditional poriferans and representatives of the Ctenophoraand Placozoa bear out this picture, then either the Hox/Para-Hox group was independently lost in the different poriferanlineages (unlikely), or animal Hox genes arose from a singlehomeoprotein ancestor in a eumetazoan ancestor (Bharathanet al., 1997).

In addition to the discovery of Hox genes, studies ofdevelopment in sea urchins, jellyfish, and hydra also providetantalizing evidence that bilateral symmetry and triploblastymay also have its origins in a common ancestor to bilateriansand cnidarians (Finnerty et al., 2004; Finnerty, 2005; Kammet al., 2006; Matus et al., 2006). It remains to be seen whetherthe Ctenophora share this ancestry. All eumetazoans share acommon developmental pathway involving the formation of

germ layers and both anterior-posterior and dorsal-ventralpatterning as the foundation for body plan formation duringembryogenesis. In contrast, the larval forms of sponges mayhave anterior-posterior organization but lack a dorsal-ventralaxis (Martindale, 2005). Thus the ultimate origin of bothHox/ParaHox genes and axial patterning may lie in theeumetazoan ancestor.

A plausible scenario of animal emergence based on the traitsof extant organisms involves a unicellular protometazoanancestor resembling a choanoflagellate and having a coloniallifestyle and a complement of cell–cell signalling and adhesiongenes. Later adaptations that evolved in a sponge-like urmeta-zoan ancestor were components of structural integrity, e.g.fibrous collagen proteins, fibronectin, and integrins (Müller,2003; Wiens et al., 2003b), caspase-based apoptosis, andsegregated germ cells. The development of a relatively simplebody plan would have been regulated by a progenitor set ofhomeoproteins; axial patterning would emerge only with theappearance of the Hox/ParaHox gene family in the eumeta-zoan ancestor. Sly et al. (2003) suggested that the ancestralform of bilaterians was a direct developer exhibiting gradualdevelopment of the adult body plan without metamorphosis.This ancestral form then developed a planktonic larva thatcould disperse and move to a wider variety of habitats, leadingto an adaptive radiation in the earliest Cambrian.

CONNECTING GEOLOGY AND GENOMES

Undoubtedly, the greatest challenge in the study of animalemergence is to connect biological evolution and geologicalchange in the late Precambrian. For example, what was theimpact of a ‘snowball Earth’ climate state on life, and is thereany connection with the appearance of the Metazoa?Complete entombment of the oceans by thick ice representsobvious metabolic challenges for life, particularly complexorganisms (Gaidos et al., 1999). Negative excursions in theisotopic record of carbon in carbonates are suggestive of,but not uniquely interpretable as, severe attenuation of thebiosphere during Neoproterozoic glaciations (Kirschvink,1992). However, the chemofossil and microfossil record doesnot support a scenario of mass extinction (Olcott et al., 2005;Corsetti et al., 2006) and, unless a much older age for theDoushantuo is accepted, the oldest unequivocal evidence foranimal life still postdates the last interval of Neoproterozoicglaciation (Narbonne & Gehling, 2003). The persistence ofice-free oases at the equator would have permitted presnowballplanktonic organisms to persist, including, presumably,the urmetazoan or eumetazoan ancestor. Runnegar (2000)speculated that blue-water refugia during a snowball eventwould have selected for organisms with extended planktoniclarval stages. Such long-lived, yolk-feeding (lecithotrophic)larvae would have included extra ‘set-aside’ cells that afterdeglaciation underwent adaptive evolution in benthic habitats(Davidson et al., 1995; Peterson & Davidson, 2000). However,

364 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Peterson (2005) has recently shown that the lecithotrophiclarval form evolved independently in multiple groups by theEarly Cambrian, perhaps as a result of selection by benthicpredation, and was not ancestral.

Atmospheric O2 is an important link between geologicalprocess and biological evolution. The appearance of abundantO2 has long been related to the development of animal life,e.g. (Nursall, 1959; Runnegar, 1982b), and new data showthat the appearance of the Ediacaran biota is contempora-neous with oxygenation of the deep ocean (Fike et al., 2006;Canfield et al., 2007). Although anoxic environments persiston the modern Earth, metazoans capable of sustained anaero-bic metabolism are invariably small and derived from aerobicancestors (Hochachka et al., 1973). In the latest of manypapers on the subject, Catling et al. (2005) argued for theuniqueness of oxygen as an ideal inorganic electron accep-tor in respiratory metabolism. They also estimated the rela-tive energy yields of anaerobic vs. aerobic metabolism ofglucose (3–12%), and the relative growth yields per energy-bearing molecule of aerobic vs. anaerobic organisms (sixtimes). They concluded that an oxygen concentration of 10%PAL was required for millimetre-sized organisms lackingcirculatory systems: higher concentrations would have beennecessary for organisms to attain centimetre sizes, and stilllarger organisms would have required complex respiratorysystems (Catling et al., 2005).

However, it is at least possible that the earliest interval ofmetazoan evolution occurred under conditions of low oxygen,perhaps a few percent PAL. An O2 level of 5% PAL by 600 Mais sufficient to explain the late Neoproterozoic sulfur isotopedata (Canfield & Teske, 1996). Flat worm-like ‘animals’ suchas Dickinsonia that required a slightly higher O2 concentration(Runnegar, 1991) and trace fossils indicative of animal activitydid not appear until about 560 Ma. Earlier organisms, includingEdiacaran Tribrachidium and rangeomorphs, may have had apassive planktonic or sessile lifestyle like those of jellyfish orsponges, one that minimized their O2 requirement. Lack of acirculation system may not have inhibited the evolution oflarge animals; sponges can grow to meter sizes, relying on asimple system of internal water circulation to provide oxygento cells. The first species may have relied exclusively on passivefilter-feeding of phytoplankton, bacterioplankton, and particularorganic matter rather than predation. More speculatively,algal symbionts could have provided an additional source ofenergy to animal hosts under conditions of low oxygen. Suchsymbionts are found in extant cnidarians (notably corals),ctenophores, acoels and at least one species of mollusc(Rumpho et al., 2000).

Whenever it happened, the mechanism(s) by which O2

levels rose to approximately the modern value is not known.O2 levels in Phanerozoic time (543 Ma to Recent) are thoughtto have been controlled by the balance between the burial oforganic matter and pyrite in marine sediments and weatheringof these reduced forms of carbon and sulfur in older, uplifted

sedimentary rocks (Lasaga & Ohmoto, 2002; Berner et al.,2003). In this picture, an increase in O2 requires either anenhancement in organic matter burial or an attenuation ofweathering. The relatively constant δ13C in carbonates overEarth history (with some exceptions as discussed above) hasbeen traditionally interpreted as indicating a constant ratio oforganic to inorganic carbon burial rate (Schidlowski, 1988).Because the total burial rate must match the total flux of CO2

from volcanoes and metamorphism of carbonate rocks, whichis assumed to be stable or only slowly decreasing, an approxi-mately constant burial rate of organic carbon is also inferred.Nevertheless, variations in δ13C are consistent with elevatedrates of organic carbon burial and a concomitant increasein O2 during the Neoproterozoic (Des Marais et al., 1992).Furthermore, Bjerrum & Canfield (2004) argued that calcu-lations of organic/inorganic burial ratio must account forremoval of inorganic carbon by hydrothermal weathering ofseafloor and that this new accounting shows a significantincrease in the burial rate of organic matter over Earth history.

Any Neoproterozoic increase in organic matter burial inturn demands a mechanistic explanation: Some scenariosinvoke the action of animals themselves to explain increasedburial efficiency, e.g. the production of faecal pellets (Loganet al., 1995), or burrowing to escape predators. The latter is acircular argument as burrowing itself probably requireselevated seawater O2; animal-independent evidence forincreasing O2 pre-dates sediment disturbance by tens ofmillion years or more (Canfield & Teske, 1996; Droser et al.,2002; Fike et al., 2006; Canfield et al., 2007). The faecalpellet scenario is tenable only if the animal gut evolved beforea benthic lifestyle (and the advent of trace fossils). Thispossibility is suggested by the acritarch record (Peterson &Butterfield, 2005), but settling of organic matter to theseafloor is only the first step in its diagenesis in marine sedi-ments and eventual preservation – or not. Searches for pelletsor organic microtextures in pre-560 Ma sedimentary rocks(Robbins et al., 1985; Krinsley et al., 1993; Brasier & McIlroy,1998) could support or rule out the faecal pellet scenario.

Changes in the cycling of other elements may have contrib-uted to the increase in oxygen. Bjerrum & Canfield (2002)proposed that the concentration of phosphorus in the lateArchean (3.8–2.4 Ga) and early Proterozoic oceans wassuppressed by adsorption onto the iron oxide particles in BIFs.Phosphorus, derived almost exclusively from weatheringof terrestrial rocks, is thought to be the limiting nutrient ongeological timescales (Benitez-Nelson, 2000) and thus low Pconcentrations would have limited net primary productivityand hence O2 production. Following the cessation of BIFdeposition, the concentration of marine P, and hence O2

production would have increased. Elevated phosphoritedeposition did occur across the Proterozoic-Phanerozoictransition (Cook & Shergold, 1984), but modern phosphoriteformation is a consequence of benthic microbial biogeochemistryat sites of ocean upwelling and not necessarily a reflection of

Emergence of animal life 365

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

seawater phosphate concentration. Adsorption on ferricoxyhydroxides may also have been moderated by competitionwith elevated levels of silica (Konhauser et al., 2007).

In contrast to phosphorus, nitrogen can be fixed from theatmosphere into a bioavailable form by some bacteria andprokaryotes and does not impose the same absolute limits onmarine productivity. However, the N-fixing enzymes nitrogenaseand nitrogenase reductase use metal cofactors such as Fe andMo, which can form insoluble sulfides under reducing con-ditions, or V, which is also depleted in anoxic ocean basins(Emerson & Huested, 1991). Deepwater anoxia in theNeoproterozoic may have removed these and other bioessentialmetals from solution, thus limiting primary production(Anbar & Knoll, 2002) (Fig. 2); oxygenation of ocean deepwater would have removed this sink. However, cyanobacterialgrowth could have been robust to changes in marine chemistry,as experiments with modern organisms suggest (Zerkle et al.,2006), by increased uptake affinity, preferential use of lessinsoluble metals, e.g. V, or changes in cellular C:N ratio. Duringlow but nonzero levels of oxygen in the Neoproterozoic,bioavailable nitrogen may also have been limited by theabsence of ammonia and depletion of nitrate by prokaryoticdenitrification (anaerobic reduction of nitrate to N2O and N2)the Neoproterozoic (Fenndel et al., 2005) (Fig. 2). The lackof fixed nitrogen limited primary production, which in turninhibited the rise of O2.

However, all of these describe positive feedbacks in theoxygen cycle, rather than triggers. In search of a trigger,attention has recently turned to terrestrial drivers of changein the marine carbon cycle during Neoproterozoic time. Lenton& Watson (2004) proposed that preferential leaching ofphosphorus from continental rocks by terrestrial fungi orlichens may have enhanced the flux of this important nutrientinto the Neoproterozoic oceans (Taunton et al., 2000; Welchet al., 2002). If marine productivity is phosphorus-limited ongeological timescales and carbon is buried with a fixed C: Pratio, an increased flux of P into the oceans would havepermitted increased organic matter burial and allowedatmospheric O2 concentrations to rise. Fungal acceleration ofsilicate weathering rates would also have lowered atmosphericCO2 and potentially initiated glaciations (Schwartzman &Volk, 1991; Heckman et al., 2001; Lenton & Watson, 2004).Kennedy et al. (2006) pointed to a Neoproterozoic shift inmarine sediment composition and a rise in radiogenic 87Sr/86Sr as evidence of increased continental weathering andtransport of clay particles to the ocean. Because clay mineralsare generally produced in biologically active soils, theyinterpret this as a signal of the proliferation of a hypotheticalterrestrial biota in the Neoproterozoic. They proposed thatclose association of organic particles with clay mineralsenhanced their preservation in marine sediments, andincreased net organic carbon burial and atmospheric O2.

One weakness of these scenarios is the absence of evidencefor a Proterozoic terrestrial biome capable of enhancing

weathering, i.e. plants or fungi. There are no undisputedland-plant body fossils or spores older than the mid-Ordovician,c. 476 Ma (Kenrick & Crane, 1997) (Fig. 1). One molecularclock analysis placed the origin of land plants at 700 Ma(Heckman et al., 2001), but, as was discussed earlier, suchestimates are highly problematic. Yuan et al. (2005) recentlyreported associations between fungal hyphae and algae orcyanobacteria in the Doushantuo formation; however, theseare thought to be of marine origin. The oldest indisputableterrestrial eukaryotic microfossils are 1.2 billion years old(Horodyski & Knauth, 1994), although there is geochemicalevidence for microbial activity in older soils (Watanabe et al.,2000; Prave, 2002; Watanabe et al., 2004), but see alsoMelezhik et al. (2004). Fungi have been shown to enhanceweathering of silicate rocks and leaching of nutrients (Gislasonet al., 1996; Brady et al., 1999; Etienne & Dupont, 2002).Searches for evidence of early land fungi, perhaps in the formof unique chemofossils, are of obvious merit. Of course, aterrestrial connection would only defer the question of whatultimately drove the timing of global change. There are noobvious environmental factors that would limit the emergenceof terrestrial biosphere to the Neoproterozoic: the smallamount of oxygen required to produce an ozone layer thatwould shield surface life against harmful UV radiation wasprobably present after the initial oxygenation event c. 2.4 Ga(Kasting, 1987).

Another concern is that increased weathering would alsohave accelerated the liberation and oxidation of reducedcarbon and sulfur from sedimentary rocks, thus enhancingthe consumption of O2. However, volcanic rocks, not sedi-mentary rocks (whose minerals have already experiencedsome degree of weathering), provide most of the Ca2+ andMg2+ ions to the oceans and drive carbonate deposition ongeological timescales. Igneous rocks, particularly basalts, arealso relatively rich in phosphorus. Preferential weathering ofthese rocks can lower atmospheric CO2 levels, cool the planet,suppress the rate of weathering of all other rock types (includingkerogen- and pyrite-containing sedimentary rocks), andthereby simultaneously increase organic matter burial anddecrease O2 consumption by weathering. In this way, weath-ering of volcanic provinces emplaced during the break-up ofRodinia and the opening of the Pacific and Iapetus oceans(Frimmel et al., 2001; Li et al., 2003) may have ushered in acolder Neoproterozoic climate regime as well as boostedoxygen levels.

It is also possible that Proterozoic O2 was set by microbialrespiration in an organic-rich ocean, rather than abiotic weath-ering of reduced carbon and sulfur on land. In this scenario,the advent of filter feeding and grazing in Ediacaran taxa assuggested by the acritarch record (Peterson & Butterfield,2005) also removed the reactive marine organic matter thatwas the substrate for microbial respiration, allowing O2 to rise.On long timescales, the rates of oxidation of volcanic gases andterrestrial reduced compounds must balance burial of organic

366 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

matter in marine sediments. But if these rates are only weaklysensitive to O2 concentration, as they may be (Betts & Holland,1991; Lasaga & Ohmoto, 2002), then the latter will bedetermined primarily by the balance between rates of primaryproduction and aerobic respiration. Over an organic-richNeoproterozoic ocean, the equality between production andconsumption would have occurred at lower pO2 because ofthe high oxygen affinity of aerobic bacteria. That large organicreservoir could have been the same reservoir of reactive carbonproposed by Rothman et al. (2003) to explain the Neoprot-erozoic oscillations in the carbon cycle. Since the advent ofanimals and the disappearance of that reservoir, the balancebetween production and consumption of oxygen has occurredat higher pO2. The transition between those states involved alower efficiency of remineralization and higher rate of carbonburial, as suggested by the carbon isotope record. Hypothet-ically, efficient filter feeding of sponge-like ancestral metazoansby 560 Ma would have produced oligotrophic conditionsin the oceans, lessening microbial activity, and allowingatmospheric O2 to rise. The virtual disappearance of suspendedorganic matter would have led to an ecological crisis thatselected for complex body plans able to predate or filter feedon larger particles (Vacelet & Duport, 2004). At the sametime, higher oxygen levels would have allowed increased bodysize and permitted colonization of the benthos, leading to theadvent of trace fossils. In this scenario, animal evolution wasboth the cause and effect of increased atmospheric oxygen.

Such a dynamic is one example of how our animal ancestorsand our planet may have coevolved through microbialintermediaries. Numerous interactions between modernanimals and microbial symbionts, commensals, and parasiteshave been documented, and there is no reason why suchinteractions were not equally well developed among theEdiacaran and Cambrian faunas (McMenamin, 1986; Fortey,2000). Microorganisms mediate many, if not most of the stepsin biogeochemical reactions important to ocean chemistry,atmospheric composition, and climate. Microbial mats domi-nated the shallow seafloor of the late Precambrian and servedas the primary habitat of benthic animals before beingoverthrown in the ‘Cambrian substrate revolution’ (Bottjer,2005). Close associations between animals (particularly sponges)and bacteria may even have provided some of the geneticsource material for metazoan evolution (Lakshminarayanet al., 2004).

Animal evolution and diversification subsequent to 555 Macould have been more strongly influenced by animal–animalinteractions than by the abiotic and microbial environment(Peterson et al., 2005; Marshall, 2006). Among many otherinnovations that accompanied the emergence of animal life,the development of the nervous system is arguably one of themost important, as it allowed the rapid receipt, processing,and transmission of information about an animal’s environ-ment and enabled complex behaviour on which natural selec-tion could have acted. Sponges have no nerve cells, but

cnidarians possess a neuronal network that includes aring-shaped concentration of cells in some taxa. The centralnervous system appears to have arisen in the common ancestorof the bilaterians (Holland, 2003) (Fig. 3). The emergence ofcomplex behaviour radically changed the environment ofanimals as they become predators, prey, mates, or competitorsof other animals (Stanley, 1976). Ecological interactionsbecome the most dynamic part of the environment, leading tothe ‘evolutionary arms race’ of further selection, specialization,and diversification that continue today.

ACKNOWLEDGEMENTS

We thank Mary Droser and Martin Kennedy for stimulatingdiscussions during a field trip to the late Precambrian and earlyCambrian units of Death Valley in California, Shuhai Xiao forhelpful discussions and copies of manuscripts in press, andVicki Pearse for extensive comments on an early version of thismanuscript. We acknowledge seven anonymous reviewers andone subject editor whose exhaustive critiques greatly improvedthis paper. EG was supported by the National Aeronauticsand Space Administration through the NASA AstrobiologyInstitute under Cooperative Agreement No. NNA04CC08A(Office of Space Science).

REFERENCES

Adoutte A, Balvoine G, Lartillot N, Lespinet O, Prud’homme B, de Rosa R (2000) The new animal phylogeny: reliability and implications. Proceedings of the National Academy of Sciences of the USA 97, 4453–4456.

Aharon P (2005) Redox stratification and anoxia of the early Precambrian oceans: implications for carbon isotope excursions and oxidation. Precambrian Research 137, 207–222.

Allen PA, Hoffman PF (2005a) Extreme winds and waves in the aftermath of a Neoproterozoic glaciation. Nature 433, 123–127.

Allen PA, Hoffman PF (2005b) Paleoclimatology: formation of Precambrian sediment ripples (reply). Nature 436, E1–E2.

Amthor JE, Grotzinger JP, Schroder S, Bowring SA, Ramezani J, Martin MW, Matter A (2003) Extinction of Cloudina and Namacalathus at the Precambrian-Cambrian boundary in Oman. Geology 31, 431–434.

Anbar AD, Knoll AH (2002) Proterozoic ocean chemistry and evolution: a bioinorganic bridge? Science 297, 1137–1142.

Aris-Brosou S, Yang Z (2003) Bayesian models of episodic evolution support a late Precambrian explosive diversification of the Metazoa. Molecular Biology and Evolution 20, 1947–1954.

Ayala FJ, Rzhetsky A, Ayala FJ (1998) Origin of the metazoan phyla: molecular clocks confirm paleontological estimates. Proceedings of the National Academy of Sciences of the USA 95, 606–611.

Bailey JV, Corsetti FA, Bottjer DJ, Marenco KN (2006) Microbially-mediated environmental influences on metazoan colonization of matground ecosystems: evidence from the lower Cambrian Harkless Formation. Palaios 21, 215–226.

Bailey JV, Joye SB, Kalanetra KM, Flood BE, Corsetti FA (2007) Evidence of giant sulphur bacteria in Neoproterozoic phosphorites. Nature 445, 198–201.

Emergence of animal life 367

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Barfod GH, Albarede F, Knoll AH, Xiao S, Frei R, Baker J (2002) New Lu-Hf and Pb-Pb age constraints on the earliest fossils. Earth and Planetary Science Letters 201, 203–212.

Bartley JK, Kah LC (2004) Marine carbon reservoir, Corg-Ccarb coupling, and the evolution of the Proterozoic carbon cycle. Geology 32, 129–132.

Bartley JK, Pope M, Knoll AH, Semikhatov MA, Petrov PY (1998) A Vendian–Cambrian boundary succession from the northwestern margin of the Siberian Platform: stratigraphy, palaeontology, chemostratigraphy and correlation. Geological Magazine 135, 473–494.

Bengtson S, Budd G (2004) Comment on ‘Small bilaterian fossils from 40 to 55 million years before the Cambrian’. Science 306, 391.

Bengtson S, Zhao Y (1992) Predatorial borings in late Precambrian mineralized exoskeletons. Science 257, 367–369.

Benitez-Nelson CR (2000) The biogeochemical cycling of phosphorus in marine systems. Earth Science Reviews 51, 109–135.

Benton MJ, Ayala FJ (2003) Dating the tree of life. Science 300, 1698–1700.

Berner RA, Beerling DJ, Dudley R, Robinson JM, Wildman RA Jr (2003) Phanerozoic atmospheric oxygen. Annual Reviews of Earth and Planetary Sciences 31, 105–134.

Betts JN, Holland HD (1991) The oxygen content of ocean bottom waters, the burial efficiency of organic carbon, and the regulation of atmospheric oxygen. Global and Planetary Change 97, 5–18.

Beukes NJ, Klein C (1992) Models for iron-formation deposition. In The Proterozoic Biosphere (eds Schopf JW, Klein C). Cambridge University Press, Cambridge, UK, pp. 147–152.

Bharathan G, Janssen B-J, Kellogg EA, Sinha N (1997) Did homeodomain proteins duplicate before the origin of angiosperms, fungi, and metazoa? Proceedings of the National Academy of Sciences of the USA 94, 13749–13753.

Bjerrum CJ, Canfield DE (2002) Ocean productivity before about 1.9 Gyr ago limited by phosphorus adsorption onto iron oxides. Nature 417, 159–162.

Bjerrum CJ, Canfield DE (2004) New insights into the burial history of organic carbon on the early Earth. Geochemistry, Geophysics, Geosystems 5, 1–9.

Blair JE, Hedges SB (2005a) Molecular clocks do not support the Cambrian explosion. Molecular Biology and Evolution 22, 387–390.

Blair JE, Hedges SB (2005b) Molecular phylogeny and divergence times of deuterostome animals. Molecular Biology and Evolution 22, 2275–2284.

Bodiselitsch B, Koeberl C, Master S, Reimold WU (2005) Estimating duration and intensity of Neoproterozoic snowball glaciations from Ir anomalies. Science 308, 239–242.

Bonner JT (2000) First Signals: the Evolution of Multicellular Development. Princeton University Press, Princeton, New Jersey.

Borchiellini C, Manuel M, Alivon E, Boury-Esnault N, Vacelet J, Le Parco Y (2001) Sponge paraphyly and the origin of Metazoa. Journal of Evolutionary Biology 14, 171–179.

Bottjer DJ (2005) Geobiology and the fossil record: eukaryotes, microbes, and their interactions. Palaeogeography, Palaeoclimatology Palaeoecology 219, 5–21.

Bowring SA, Myrow P, Landing E, Ramezani J, Grotzinger J (2003) Geochronological constraints on terminal Neoproterozic events and the rise of metazoans. Geophysical Research Abstracts 5, 13219.

Boyce M, Degterev A, Yuan J (2004) Caspases: an ancient cellular sword of Damocles. Cell Death and Differentiation 11, 29–37.

Brady PV, Dorn RI, Brazel AJ, Clark J, Moore RB, Glidewell T (1999) Direct measurement of the combined effects of lichen, rainfall, and temperature on silicate weathering. Geochimica et Cosmochimica Acta 63, 3293–3300.

Brasier MD, Antcliffe J (2004) Decoding the Ediacaran enigma. Science 305, 1115–1117.

Brasier MD, Green O, Shields G (1997) Ediacaran sponge spicules from southwestern Mongolia and the origins of the Cambrian fauna. Geology 25, 303–306.

Brasier MD, McIlroy D (1998) Neonereites uniserialis from c. 600 Ma year old rocks in western Scotland and the emergence of animals. Journal of the Geological Society 155, 5–12.

Brocks JJ, Logan GA, Buick R, Summons RE (1999) Archean molecular fossils and the early rise of eukaryotes. Science 285, 1033–1036.

Brocks JJ, Love GD, Summons RE, Knoll AH, Logan GA, Bowden SA (2005) Biomarker evidence for green and purple sulphur bacteria in a stratified Palaeoproterozoic sea. Nature 437, 866–870.

Bromham L (2003) What can DNA tell us about the Cambrian explosion? Integrative and Comparative Biology 43, 148–156.

Bromham L (2006) Molecular dates for the Cambrian explosion: is the light at the end of the tunnel an oncoming train? Palaeontologia Electronica 9(1), 2e.

Bromham LD, Hendy MD (2000) Can fast early rates reconcile molecular dates with the Cambrian explosion? Proceedings of the Royal Society of London. Series B: Biology Sciences 267, 1041–1047.

Brooke NM, Garcia-Fernández J, Holland PWH (1998) The ParaHox gene cluster is an evolutionary sister of the Hox gene cluster. Nature 392, 920–922.

Brusca RC, Brusca GJ (2003) Invertebrates, Sinnauer Associates, Sunderland, Massachusetts.

Budd GE, Jensen S (2000) A critical reappraisal of the fossil record of the bilaterian phyla. Biology Reviews of the Cambridge Philosophical Society 75, 253–295.

Buss LW (1999) Slime molds, ascidians, and the utility of evolutionary theory. Proceedings of the National Academy of Sciences of the USA 96, 8801–8803.

Buss LW, Seilacher A (1994) The phylum Vendobionta: a sister group of the eumetazoa? Paleobiology 20, 1–4.

Butterfield NJ (1994) Burgess Shale-type fossils from a Lower Cambrian shallow-shelf sequence in northwestern Canada. Nature 369, 477–479.

Butterfield NJ (1997) Plankton ecology and the Proterozoic–Phanerozoic transition. Paleobiology 23, 247–262.

Butterfield NJ (2000) Bangiomorpha pubescens n. gen. n. sp.: implications for the evolution of sex, multicellularity, and the Mesoproterozoic/Neoproterozoic radiation of eukaryotes. Paleobiology 2000, 386–404.

Butterfield NJ (2005) Probable Proterozoic fungi. Paleobiology 31, 165–182.

Butterfield NJ (2007) Macroevolution and macroecology through deep time. Journal of Paleontology 50, 41–55.

Calver CR, Black LP, Everard JL, Seymour DB (2004) U-Pb zircon age constraints on late Neoproterozoic glaciation in Tasmania. Geology 32, 893–896.

Canfield DE (1998) A new model for Proterozoic ocean chemistry. Nature 396, 450–453.

Canfield DE (2005) The early history of atmospheric oxygen: homage to Robert M. Garrels. Annual Reviews of Earth and Planetary Sciences 33, 1–36.

Canfield DE, Poulton SW, Narbonne GM (2007) Late-Neoproterozoic deep-ocean oxygen and the rise of animal life. Science 315, 92–94.

Canfield DE, Raiswell R (1999) The evolution of the sulfur cycle. American Journal of Science 299, 697–723.

Canfield DE, Teske A (1996) A Late Proterozoic rise in atmospheric oxygen concentration inferred from phylogenetic and sulfur-isotope studies. Nature 382, 127–132.

368 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Catling DC, Glein CR, Zahnle KJ, McKay CP (2005) Why O2 is required by complex life on habitable planets and the concept of planetary ‘oxygenation time’. Astrobiology 5, 415–438.

Chen J-Y, Bottjer DJ, Oliveri P, Dornbos SQ, Gao F, Ruffins S, Chi H, Li C-W, Davidson EH (2004b) Small bilaterian fossils from 40 to 55 million years before the Cambrian. Science 305, 218–222.

Chen DF, Dong WQ, Zhu BQ, Chen XP (2004a) Pb-Pb ages of Neoproterozoic phosphorites in South China: constraints on early metazoan evolution and glaciation events. Precambrian Research 132, 123–132.

Chen J-Y, Oliveri P, Davidson E, Bottjer DJ (2004c) Response to comment on ‘Small bilaterian fossils from 40 to 55 million years before the Cambrian’. Science 306, 1291.

Chen J-Y, Oliveri P, Gao F, Dornbos SQ, Li C-W, Bottjer DJ, Davidson EH (2002) Precambrian animal life: probable developmental and adult cnidarian forms from China. Developmental Biology 248, 182–196.

Chen J-Y, Oliveri P, Li C-W, Zhou G-Q, Gao F, Hagadorn JW, Peterson KJ, Davidson EH (2000) Precambrian animal diversity: putative phosphatized embryos from the Doushantuo Formation of China. Proceedings of the National Academy of Sciences of the USA 97, 4457–4462.

Chourrout D, Delsuc F, Chourrout P, Edvardsen, RB, Rentzsch, F, Renfer, E, Jensen, MF, Zhu, B, de Jong, P, Steele, RE, others (2006) Minimal ProtoHox cluster inferred from bilaterian and cnidarian Hox complements. Nature 442, 684–687.

Clapham ME, Narbonne GM, Gehling JG (2003) Paleoecology of the oldest known animal communities: Ediacaran assemblages at Mistaken Point, Newfoundland. Paleobiology 29, 527–544.

Colodner DC, Boyle EA, Edmond JM, Thomson J (1992) Post-depositional mobility of platinum, iridium and rhenium in marine sediments. Nature 358, 402–404.

Condon D, Zhu M, Bowring SA, Wang W, Yang A, Jin Y (2005) U-Pb ages from the Neoproterozoic Doushantuo formation, China. Science 308, 95–98.

Conway Morris S (2002) Ancient animals or something else entirely? Science 298, 57–58.

Conway Morris S (2006) Darwin’s dilemma: the realities of the Cambrian ‘explosion’. Philosophical Transactions of the Royal Society of London. Series B: Biology Sciences 361, 1069–1083.

Cook PJ, Shergold JH (1984) Phosphorus, phosphorites and skeletal evolution at the Precambrain–Cambrian boundary. Nature 308, 231–236.

Corsetti FA, Olcott AN, Bakermans C (2006) The biotic response to Neoproterozoic snowball Earth. Palaeogeography, Palaeoclimatology, Palaeoecology 232, 114–130.

Davidson EH, Erwin DH (2006) Gene regulatory networks and the evolution of animal body plans. Science 311, 796–800.

Davidson EH, Peterson KJ, Cameron RA (1995) Origin of bilaterian body plans: evolution of developmental regulatory mechanisms. Science 270, 1319–1325.

Dellaporta SL, Xu A, Sagasser S, Jakob W, Moreno MA, Buss LW, Schierwater B (2006) Mitochondrial genome of Trichoplax adhaerens supports Placozoa as the basal lower metazoan phylum. Proceedings of the National Academy of Sciences of the USA 103, 8751–8756.

Des Marais DJ, Strauss H, Summons RE, Hayes JM (1992) Carbon isotope evidence for the stepwise oxidation of the Proterozoic environment. Nature 359, 605–609.

Donnadieu Y, Goddéris Y, Ramstein G, Nédélec A, Meert J (2004) A ‘snowball Earth’ climate triggered by continental break-up through changes in runoff. Nature 428, 303–306.

Dornbos SQ, Bottjer DJ, Chen J-Y, Gao F, Oliveri P, Li C-W (2006) Environmental controls on the taphonomy of phosphatized

animals and animal embryos from the Neoproterozoic Doushantuo formation, southwest China. Palaios 21, 3–14.

Douzery EJP, Snell EA, Babteste E, Delsuc F, Philippe H (2004) The timing of eukaryotic evolution: does a relaxed molecular clock reconcile proteins and fossils? Proceedings of the National Academy of Sciences of the USA 101, 15386–15391.

Droser ML, Gehling JG, Jensen S (1999) When the worm turned: coincidence of early Cambrian ichnofabric and trace-fossil record in silicoclastic rocks of South Australia. Geology 27, 625–628.

Droser ML, Jensen S, Gehling JG (2002) Trace fossils and substrates of the terminal Proterozoic–Cambrian transition: Implications for the record of early bilaterians and sediment mixing. Proceedings of the National Academy of Sciences of the USA 99, 12572–12576.

Dzik J (1999) Organic membranous skeleton of the Precambrian metazoans from Namibia. Geology 27, 519–522.

Dzik J (2002) Possible ctenophoran affinities of the Precambrian ‘Sea-Pen’ Rangea. Journal of Morphology 252, 315–334.

Eichinger L, Pachebat JA, Glöckner G, Rajandrream M-A, Sucgang R, Berriman M, Song J, Olsen R, Szafranski K, Xu Q, Tunggal B, Kummerfeld S, Madera M, Konfortov BA, Rivero F, Bankier AT, Lehmann R, Hamlin N, Davies R, Gaudet P, Fey P, Pilcher K, Chen G, Saunders D, Sodergren E, Davis P, Kerhornou A, Nie X, Hall N, Anjard C, Hemphill L, Bason N, Farbrother P, Desany B, Just E, Morio T, Rost R, Churcher C, Cooper J, Haydock S, van Driessche N, Cronin A, Goodhead I, Muzny D, Mourier T, Pain A, Lu M, Harper D, Lindsay R, Hauser H, James K, Quiles M, Madan Babu M, Saito T, Buchrieser C, Wardroper A, Felder M, Thangavelu M, Johnson D, Knights A, Loulseged H, Mungall K, Oliver K, Price C, Quail MA, Urushihara H, Hernandez J, Rabbinowitsch E, Steffen D, Sanders M, Ma J, Kohara Y, Sharp S, Simmonds M, Spiegler S, Tivey A, Sugano S, White B, Walker D, Woodward J, Winckler T, Tanaka Y, Shaulsky G, Schleicher M, Weinstock G, Rosenthal A, Cox EC, Chisholm RL, Gibbs R, Loomis WF, Platzer M, Kay RR, Williams J, Dear PH, Noegel AA, Barrell B, Kuspa A (2005) The genome of the social amoeba Dictyostelium discoideum. Nature 435, 43–57.

Emerson SR, Huested SS (1991) Ocean anoxia and the concentrations of molybdenum and vanadium in seawater. Marine Chemistry 34, 177–196.

Erwin DH (1999) The origin of bodyplans. American Zoologist 39, 617–629.

Etienne S, Dupont J (2002) Fungal weathering of basaltic rocks in a cold oceanic environment (Iceland): comparison between experimental and field observations. Earth Surface Processes and Landforms 27, 737–748.

Extavour CG, Akam M (2003) Mechanisms of germ cell specification across the metazoans: epigenesis and preformation. Development 130, 5869–5884.

Eyles N, Januszczak N (2004) ‘Zipper-rift’: a tectonic model for Neoproterozoic glaciations during the breakup of Rodinia after 750 Ma. Earth Science Reviews 65, 1–73.

Fanning CM, Link PK (2004) U-Pb SHRIMP ages of Neoproterozoic (Sturtian) glaciogenic Pocatello Formation, southeastern Idaho. Geology 32, 881–884.

Fedonkin MA (2003) The origin of the Metazoa in the light of the Proterozoic fossil record. Paleontological Research 7, 9–41.

Fenndel K, Follows M, Falkowski PG (2005) The co-evolution of the nitrogen, carbon, and oxygen cycles in the Proterozoic ocean. American Journal of Science 305, 526–545.

Fike DA, Grotzinger J, Pratt LM, Summons RE (2006) Oxidation of the Ediacaran ocean. Nature 444, 744–747.

Finnerty JR (2005) Did internal transport, rather than directed locomotion, favor the evolution of bilateral symmetry in animals? Bioessays 27, 1174–1180.

Emergence of animal life 369

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Finnerty JR, Martindale MQ (1999) Ancient origins of axial patterning genes: Hox genes and ParaHox genes in the Cnidaria. Evolution and Development 1, 16–23.

Finnerty JR, Master VA, Irvine S, Kourakis MJ, Warriner S, Martindale MQ (1996) Homeobox genes in the Ctenophora: identification of paired-type and Hox homologues in the atentaculate ctenophore, Beroe ovata. Molecular Marine Biology and Biotechnology 5, 249–258.

Finnerty JR, Pang K, Burton P, Paulson D, Martindale MQ (2004) Origins of bilateral symmetry: Hox and Dpp expression in a sea anemone. Science 304, 1335–1337.

Fortey RA (2000) Olenid trilobites: the oldest known chemoautotrophic symbionts? Proceedings of the National Academy of Sciences of the USA 97, 6574–6578.

Frimmel HE, Zartman RE, Späth A (2001) The Richtersveld igneous complex, South Africa: U-Pb zircon and geochemical evidence for the beginning of Neoproterozoic continental breakup. Journal of Geology 109, 493–508.

Fulton C (1993) Naegleria – a research partner for cell and developmental biology. Journal of Eukaryotic Microbiology 40, 520–532.

Gaidos EJ, Nealson KH, Kirschvink JL (1999) Life in ice-covered oceans. Science 284, 1631–1633.

Garcia-Fernández J (2005a) The genesis and evolution of homeobox gene clusters. Nature Reviews Genetics 6, 881–892.

Garcia-Fernández J (2005b) Hox, ParaHox, ProtoHox: facts and guesses. Heredity 94, 145–152.

Gellon G, McGinnis W (1998) Shaping animal body plans in development and evolution by modulation of Hox expression patterns. Bioessays 120, 116–125.

Gilbert SF (2003) Opening Darwin’s black box: teaching evolution through developmental genetics. Nature Reviews Genetics 4, 735–741.

Gislason SR, Arnornsson S, Armannsson H (1996) Chemical weathering of basalt in Southwest Iceland: effects of runoff, age of rocks and vegetative/glacial cover. American Journal of Science 296, 837–907.

Glenner H, Hansen AJ, Sorensen MV, Ronquist F, Huelsenbeck JP, Willerslev E (2004) Bayesian inference of the metazoan phylogeny, a combined molecular and morphological approach. Current Biology 14, 1644–1649.

Gomer RH (2001) Not being the wrong size. Nature Reviews Molecular and Cellular Biology 2, 48–54.

Goodman CS, Coughlin BC (2000) The evolution of evo-devo biology. Proceedings of the National Academy of Sciences of the USA 97, 4424–4425.

Goodman JC, Pierrehumbert RT (2003) Glacial flow of floating marine ice in ‘Snowball Earth’. Journal of Geophysical Research 108, 3308.

Gough DO (1981) Solar interior structure and luminosity variations. Solar Physics 74, 21–34.

Graur D, Martin W (2004) Reading the entrails of chickens: molecular timescales of evolution and the illusion of precision. Trends in Genetics 20, 80–86.

Grey K, Walter MR, Calver CR (2003) Neoproterozoic biotic diversification: snowball Earth or aftermath of the Acraman impact? Geology 31, 459–462.

Grice K, Cao C, Love GD, Bottjer ME, Twitchett RJ, Grosjean E, Summons RE, Turgeon SC, Dunning W, Jin Y (2005) Photic zone euxinia during the Permian-Triassic superanoxic event. Science 307, 706–709.

Grotzinger JP, Kasting JF (1993) New constraints on Precambrian ocean composition. Journal of Geology 101, 235–243.

Grotzinger J, Knoll AH (1995) Anomalous carbonate precipitates: is the Precambrian the key to the Permian? Palaios 10, 578–596.

Grotzinger JP, Watters WA, Knoll AH (2000) Calcified metazoans in thrombolite–stromatolite reefs of the terminal Proterozoic Nama Group, Namibia. Paleobiology 26, 334–359.

Hagadorn JW, Xiao S, Donoghue PCJ, Bengtson S, Gostling NJ, Pawlowska M, Raff EC, Raff RA, Turner FR, Chongyu Y, others (2006) Cellular and subcellular structure of Neoproterozoic animal embryos. Science 314, 291–294.

Halanych KM (2004) The new view of animal phylogeny. Annual Reviews of Ecology, Evolution, and Systematics 35, 229–256.

Halverson GP, Hoffman PF, Schrag DP, Kaufman AJ (2002) A major perturbation of the carbon cycle before the Ghaub glaciation (Neoproterozoic) in Namibia: Prelude to snowball Earth? Geochemistry, Geophysics, Geosystems 3, 10.1029/2001GC000244.

Hannon GJ (2003) RNA interference. Nature 418, 244–251.Harland WB (1964) Evidence of late Precambrian glaciation and its

significance. In Problems in Paleoclimatology (ed. Nairn AEM). John Wiley & Sons, London, pp. 119–149.

Hayes JM (2001) Fractionation of carbon and hydrogen isotopes in biosynthetic processes. Reviews of Mineralogy and Geochemistry 43, 225–277.

Hazkani-Covo E, Levanon YE, Rotman G, Graur D, Novik A (2004) Evolution of multicellularity in Metazoa: comparative analysis of the subcellular localization of proteins in Saccharomyces, Drosophila and Caenorhabditis. Cell Biology International 28, 171–178.

Heckman DS, Geiser DM, Eidell BR, Stauffer RL, Kardos NL, Hedges SB (2001) Molecular evidence for the early colonization of land by fungi and plants. Science 293, 1129–1133.

Ho SYW, Phillips MJ, Cooper A, Drummond AJ (2005) Time dependency of molecular rate estimates and systematic overestimation of recent divergence times. Molecular Biology and Evolution 22, 1561–1568.

Hochachka PW, Fields J, Mustafa T (1973) Animal life without oxygen: basic biochemical mechanisms. Integrative and Comparative Biology 13, 543–555.

Hoffman PF (1999) The break-up of Rodinia, birth of Gondwana, true polar wander and the snowball Earth. Journal of African Earth Science 28, 17–33.

Hoffman PF, Kaufman AJ, Halverson GP, Schrag DP (1998) A Neoproterozoic snowball Earth. Science 281, 1342–1346.

Hoffman PF, Schrag DP (2002) The snowball Earth hypothesis: testing the limits of global change. Terra Nova 14, 129–155.

Hoffmann KH, Condon DJ, Bowring SA, Crowley TJ (2004) U-Pb zircon date from the Neoproterozoic Ghaub formation, Namibia: Constraints on Marinoan glaciation. Geology 32, 817–820.

Hofmann HJ, Narbonne GM, Aiken JD (1990) Ediacaran remains from intertillite beds in northwestern Canada. Geology 18, 1199–1202.

Holland ND (2003) Early central nervous system evolution: an era of skin brains? Nature Reviews Genetics 4, 1–11.

Horodyski RJ, Knauth LP (1994) Life on land in the Precambrian. Science 263, 494–498.

Hua H, Pratt BR, Zhang L-Y (2003) Borings in Cloudina shells: complex predator–prey dynamics in the terminal Neoproterozoic. Palaios 18, 454–459.

Huelsenbeck JP, Ronquist F, Nielsen C, Bollback JP (2001) Bayesian inference of phylogeny and its impact on evolutionary biology. Science 294, 2310–2314.

Huntley JW, Xiao S, Kowalewski M (2006) 1.3 billion years of acritarch history: an empirical morphospace approach. Precambrian Research 144, 52–68.

Hyde WT, Crowley TJ, Baum SK, Peltier WR (2000) Neoproterozoic ‘snowball Earth’ simulations with a coupled climate/ice-sheet model. Nature 405, 425–429.

Hyman LH (1940) The Invertebrates. Protozoa Through Ctenophora. McGraw-Hill, New York.

370 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Ivantsov AY, Fedonkin MA (2002) Conulariid-like fossil from the Vendian of Russia: a metazoan clade across the Proterozoic/Palaeozoic boundary. Palaeontology 45, 1219–1229.

Jaanusson V (1976) Faunal dynamics in the Middle Ordovician (Viruan) of Balto-Scandia. In The Ordovician System: Proceedings of a Paleontological Association Symposium (ed. Bassett MG). University of Wales Press and National Museum, Wales, Cardiff, pp. 301–326.

Jakob W, Sagasser S, Dellaporta SL, Holland P, Kuhn K, Schierwater B (2004) The Trox-2 Hox/ParaHox gene of Trichoplax (Placozoa) marks an epithelial boundary. Developmental Genes and Evolution 214, 170–175.

James NP, Narbonne GM, Dalrymple RW, Kyser TK (2005) Glendonites in Neoproterozoic low-latitude, interglacial, sedimentary rocks, northwest Canada: insights into the Cryogenian ocean and Precambrian cold-water carbonates. Geology 23, 9–12.

Javaux EJ, Knoll AH, Walter MR (2001) Morphological and ecological complexity in early eukaryotic ecosystems. Nature 412, 66–69.

Jensen S (2003) The Proterozoic and earliest Cambrian trace fossil record: patterns, problems, and perspectives. Integrative and Comparative Biology 43, 219–228.

Jensen S, Droser ML, Gehling JG (2005) Trace fossil preservation and the early evolution of animals. Palaeogeography, Palaeoclimatology, Palaeoecology 220, 19–29.

Jensen S, Saylor BZ, Gehling JG, Germs GJB (2000) Complex trace fossils from the terminal Neoproterozoic of Namibia. Geology 28, 143–146.

Jermiin LS, Poladian L, Charleston MA (2005) Is the ‘big bang’ in animal evolution real? Science 310, 1910–1911.

Jerolmack DJ, Mohrig D (2005) Paleoclimatology: formation of Precambrian sediment ripples. Nature 436, E1.

Kah LC, Lyons TW, Frank TD (2004) Low marine sulphate and protracted oxygenation of the Proterozoic biosphere. Nature 431, 834–837.

Kaiser D (2001) Building a multicellular organism. Annual Reviews of Genetics 35, 103–123.

Kamm K, Schierwater B, Jakob W, Dellaporta SL, Miller DJ (2006) Axial patterning and diversification in the cnidaria predate the Hox system. Current Biology 16, 1–7.

Kappler A, Pasquero C, Konhauser KO, Newman DK (2005) Deposition of banded iron formations by anoxygenic phototrophic Fe (II) -oxidizing bacteria. Geology 33, 865–868.

Kasting JF (1987) Theoretical constrains on oxygen and carbon dioxide concentrations in the Precambrian atmosphere. Precambrian Research 34, 305–329.

Kasting JF, Catling D (2003) Evolution of a habitable planet. Annual Reviews of Astronomy and Astrophysics 41, 429–463.

Kaufman AJ (2005) The calibration of Ediacaran time. Science 308, 59–60.

Kaufman AJ, Xiao S (2003) High CO2 levels in the Proterozoic atmosphere estimated from analyses of individual microfossils. Nature 425, 279–282.

Kennedy MJ, Christie-Black N, Sohl LE (2001) Are Proterozoic cap carbonates and isotopic excursions a record of gas hydrate destabilization following Earth’s coldest intervals? Geology 29, 443–446.

Kennedy M, Droser M, Mayer LM, Pevear D, Mrofka D (2006) Late Precambrian oxygenation; inception of the clay mineral factory. Science 311, 1446–1449.

Kennedy MJ, Runnegar B, Prave AR, Hoffmann KH, Arthur MA (1998) Two or four Neoproterozoic glaciations? Geology 26, 1059–1063.

Kenrick P, Crane PR (1997) The origin and early evolution of plants on land. Nature 389, 33–39.

King N (2004) The unicellular ancestry of animal development. Developmental Cell 7, 313–325.

King N, Hittinger CT, Carroll SB (2003) Evolution of key cell signalling and adhesion protein families predates animal origins. Science 301, 361–363.

Kirk LD (2005) A twelve-step program for evolving multicellularity and a division of labor. Bioessays 27, 299–310.

Kirschvink JL (1992) Late Proterozoic low-latitude global glaciation: the snowball Earth. In The Proterozoic Biosphere (eds Schopf JW, Klein C). Cambridge University Press, Cambridge, pp. 51–52.

Kirschvink JL, Raub TD (2003) A methane fuse for the Cambrian explosion: carbon cycles and true polar wander. Comptes Rendus Geoscience 335, 65–78.

Klein C (2005) Some Precambrian banded iron-formations (BIFs) from around the world: their age, geologic setting, mineralogy, metamorphism, geochemistry, and origin. American Mineraologist 90, 1473–1499.

Knoll AH (1994) Proterozoic and Early Cambrian protists: evidence for accelerating evolutionary tempo. Proceedings of the National Academy of Sciences of the USA 91, 6743–6750.

Knoll AH, Carroll SB (1999) Early animal evolution: emerging views from comparative biology and geology. Science 284, 2129–2137.

Knoll AH, Walter MR, Narbonne GM, Christie-Blick N (2004) A new period for the geologic time scale. Science 305, 621–622.

Konhauser KO, Lalonde SV, Amskold L, Holland HD (2007) Was there really an Archean phosphate crisis. Science 315, 1234.

Koonin EV, Aravind L (2002) Origin and evolution of eukaryotic apoptosis: the bacterial connection. Cell Death and Differentiation 9, 394–404.

Krinsley D, Nagy B, Dypvik H, Rigali M (1993) Microtextures in mudrocks as revealed by backscattered electron imaging. Precambrian Research 61, 191–207.

Kumar S (1995) Megafossils from the Mesoproterozoic Rohtas Formation (the Vindhyan Supergroup), Katni area, central India. Precambrian Research 72, 171–184.

Kump L (1991) Interpreting carbon-isotope excursions: Strangelove oceans. Geology 19, 299–302.

Lakshminarayan MI, Aravind L, Coon SL, Klein DC, Koonin EV (2004) Evolution of cell-cell signaling in animals: did late horizontal gene transfer from bacteria have a role? Trends in Genetics 20, 292–299.

Landing E (1994) Precambrian–Cambrian boundary global stratotype ratified and a new perspective of Cambrian time. Geology 22, 179–182.

Larroux C, Fahey B, Liubicich D, Hinman VF, Gauthier M, Gongora M, Green K, Worheide G, Leys SP, Degnan BM (2006) Developmental expression of transcription factor genes in a demosponge: insights into the origin of metazoan multicellularity. Evolution and Development 8, 150–173.

Lasaga AC, Ohmoto H (2002) The oxygen geochemical cycle: dynamics and stability. Geochimica et Cosmochimica Acta 66, 361–381.

Lavrov DV, Forget L, Kelly M, Lang BF (2005) Mitochondrial genomes of two demosponges provide insights into an early stage of animal evolution. Molecular Biology and Evolution 22, 1231–1239.

Laws EA, Bidigare RR, Popp BN (1997) Effect of growth rate and CO2 concentration on carbon isotopic fractionation by the marine diatom Phaeodactylum tricornutum. Limnology and Oceanography 42, 1552–1560.

Lenton TM, Watson AJ (2004) Biotic enhancement of weathering, atmospheric oxygen and carbon dioxide in the Neoproterozoic. Geophysical Research Letters 31, L05202.

Levinton J, Dubb L, Wray G (2004) Simulations of evolutionary

Emergence of animal life 371

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

radiations and their application to understanding the probability of a Cambrian explosion. Journal of Paleontology 78, 31–38.

Li C-W, Chen J-Y, Hua H (1998a) Precambrian sponges with cellular structures. Science 279, 879–882.

Li C-W, Chen J-Y, Hua T-E (1998b) Interpreting late Precambrian microfossils. Science 282, 1783–1784.

Li ZX, Li XH, Kinny PD, Wang J, Zhang S, Zhou H (2003) Geochronology of Neoproterozoic syn-rift magmatism in the Yangtze Craton, South China and correlations with other continents: evidence for a mantle superplume that broke up Rodinia. Precambrian Research 122, 85–109.

Littler MM, Littler DS, Blair SM, Norris JN (1985) Deepest known plant life discovered on an uncharted seamount. Science 227, 57–59.

Logan GA, Hayes JM, Hieshima GB, Summons RE (1995) Terminal Proterozoic reorganization of biogeochemical cycles. Nature 376, 53–56.

Logan GA, Summons RE, Hayes JM (1997) An isotopic biogeochemical study of Neoproterozoic and early Cambrian sediments from the Centralian Superbasin, Australia. Geochimica et Cosmochimica Acta 61, 5391–5409.

Maldonado M (2004) Choanoflagellates, choanocytes, and animal multicellularity. Invertebrate Biology 123, 1–22.

Marshall CR (2006) Explaining the Cambrian ‘explosion’ of animals. Annual Reviews of Earth and Planetary Sciences 34, 355–384.

Martin MW, Grazhdankin DV, Bowring SA, Evans DAD, Fedonkin MA, Kirschvink JL (2000) Age of Neoproterozoic bilaterian body and trace fossils, White Sea, Russia: implications for metazoan evolution. Science 288, 841–845.

Martindale MQ (2005) The evolution of metazoan axial properties. Nature Reviews Genetics 6, 917–927.

Martindale MQ, Pang K, Finnerty JR (2004) Investigating the origins of triploblasty: ‘mesodermal’ gene expression in a diploblastic animal, the sea anemone Nematostella vectensis (phylum, Cnidaria; class, Anthozoa). Development 131, 2463–2474.

Martinelli C, Spring J (2004) Expression pattern of the homeobox gene Not in the basal metazoan Trichoplax adhaerens. Gene Expression Patterns 4, 443–447.

Matus DQ, Pang K, Marlow H, Dunn CW, Thomsen GH, Martindale MQ (2006) Molecular evidence for deep evolutionary roots of bilaterality in animal development. Proceedings of the National Academy of Sciences of the USA 103, 11195–11200.

McMenamin MAS (1986) The garden of Ediacara. Palaios 1, 178–182.

McNaughton RB, Narbonne GM (1999) Evolution and ecology of Neoproterozoic-lower Cambrian trace fossils, NW Canada. Palaios 14, 97–115.

Medina M, Collins AG, Silberman JD, Sogin ML (2001) Evaluating hypotheses of basal animal phylogeny using complete sequences of large and small subunit rRNA. Proceedings of the National Academy of Sciences of the USA 98, 9707–9712.

Melezhik VA, Fallick AE, Grillo SM (2004) Subaerial exposure surfaces in a Paleoproterozoic C-13-rich dolostone sequence from the Pechaenga Greenstone belt: palaeoenvironmental and isotopic implications for the 2330–2060 Ma global isotope excursion of C-13/C-12. Precambrian Research 133, 75–103.

Mendoza L, Taylor JW, Ajello L (2002) The class Mesomycetozoea: a heterogeneous group of microorganisms at the animal–fungal boundary. Annual Reviews of Microbiology 56, 315–344.

Müller WEG (2003) The origin of metazoan complexity: porifera as integrated animals. Integrative and Comparative Biology 43, 3–10.

Narbonne GM (2005) The Ediacara biota: Neoproterozoic origin of animals and their ecosystems. Annual Reviews of Earth and Planetary Sciences 33, 421–442.

Narbonne GM, Gehling JG (2003) Life after snowball: the oldest complex Ediacaran fossils. Geology 31, 27–30.

Nichols SA, Dirks W, Pearse JS, King N (2006) Early evolution of animal cell signaling and adhesion genes. Proceedings of the National Academy of Sciences of the USA 103, 12451–12456.

Nielsen C, Martinez P (2003) Patterns of gene expression: homology or homocracy? Developmental Genes and Evolution 213, 149–154.

Nursall JR (1959) Oxygen as a prerequisite to the origin of the Metazoa. Nature 183, 1170–1172.

Olcott AN, Sessions AL, Corsetti FA, Kaufman AJ, Flavio de Oliviera T (2005) Biomarker evidence for photosynthesis during Neoproterozoic glaciation. Science 310, 471–474.

Pavlov AA, Hurtgen MT, Kasting JF, Arthur MA (2003) Methane-rich Proterozoic atmosphere? Geology 31, 87–90.

Pereira SL, Baker AJ (2006) A mitogenomic timescale for birds detects variable phylogenetic rates of molecular evolution and refutes the standard molecular clock. Molecular Biology and Evolution 23, 1731–1740.

Perovic S, Schröder HC, Sudek S, Grebenjuk VA, Batel R, Stifanic M, Müller IM, Müller WEG (2003) Expression of one sponge Iroquois homeobox gene in primmorphs from the Suberites domuncula during canal formation. Evolution and Development 5, 240–250.

Peterson KJ (2005) Macroevolutionary interplay between planktic larvae and benthic predators. Geology 33, 929–932.

Peterson KJ, Butterfield NJ (2005) Origin of the eumetazoa: testing ecological predictions of molecular clocks against the proterozoic fossil record. Proceedings of the National Academy of Sciences of the USA 102, 9547–9552.

Peterson KJ, Davidson EH (2000) Regulatory evolution and the development of the bilaterians. Proceedings of the National Academy of Sciences of the USA 97, 4430–4433.

Peterson KJ, Lyons JB, Nowak KS, Takacs CM, Wargo MJ, McPeek MA (2004) Estimating metazoan divergence times with a molecular clock. Proceedings of the National Academy of Sciences of the USA 101, 6536–6541.

Peterson KJ, McPeek MA, Evans DAD (2005) Tempo and mode of early animal evolution: inferences from rocks, Hox, and molecular clocks. Paleobiology 31, S36–S55.

Peucker-Ehrenbrink B, Ravizza G (2000) The marine osmium isotope record. Terra Nova 12, 205–219.

Pierrehumbert RT (2004) High levels of atmospheric carbon dioxide necessary for the termination of global glaciation. Nature 429, 646–649.

Pisani D, Poling LL, Lyons-Weiler M, Hedges SB (2004) The colonization of land by animals: molecular phylogeny and divergence times among arthropods. BMC Evolutionary Biology 2, 1–10.

Pollard D, Kasting JF (2005) Snowball Earth: a thin-ice solution with flowing sea glaciers. Journal of Geophysical Research 110, 10.1029/2004JC002525.

Porter SM, Knoll AH (2000) Testate amoebae in the Neoproterooic Era: evidence from vase-shaped microfossils in the Chuar Group, Grand Canyon. Paleobiology 26, 360–385.

Porter SM, Knoll AH (2003) Vase-shaped microfossils from the Neoproterozoic Chuar Group, Grand Canyon: a classification guided by modern testate amoebae. Journal of Paleontology 77, 409–429.

Poulton SW, Fralick PW, Canfield DE (2004) The transition to a sulphidic ocean ~1.84 billion years ago. Nature 431, 173–177.

Prave AR (2002) Life on land in the Proterozoic: evidence from the Torridonian rocks of northern Scotland. Geology 30, 811–814.

Raff RA, Villinski JT, Tuner FR, Donoghue PCJ, Raff RA (2006) Experimental taphonomy shows the feasibility of fossil embryos.

372 E. GAIDOS et al.

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Proceedings of the National Academy of Sciences of the USA 103, 5846–5851.

Rai V, Gautam R (1999) Evaluating evidence of ancient animals. Science 284, 1235.

Rasmussen B, Bengtson S, Fletcher IR, McNaughton NJ (2002a) Ancient animals or something else entirely. Science 298, 58–59.

Rasmussen B, Bengtson S, Fletcher IR, McNaughton NJ (2002b) Discoidal impressions and trace-like fossils more than 1200 million years old. Science 296, 1112–1115.

Rasmussen B, Bose PK, Sarkar S, Banerjee S, Fletcher IR, McNaughton NJ (2002c) 1.6 Ga U-Pb zircon age for the Chorhat Sandstone, lower Vindhyan, India: possible implications for early evolution of animals. Geology 30, 103–106.

Redecker D, Kodner R, Graham LE (2000) Glomalean fungi from the Ordovician. Science 289, 1920–1921.

Ridgwell AJ, Kennedy MJ, Caldeira K (2003) Carbonate deposition, climate stability, and Neoproterozoic ice ages. Science 302, 859–862.

Robbins EI, Porter KG, Haberyan KA (1985) Pellet microfossils: possible evidence for metazoan life in Early Proterozoic time. Proceedings of the National Academy of Sciences of the USA 82, 5809–5913.

Roger AJ, Hug LA (2006) The origin and diversification of eukaryotes: problems with molecular phylogenetics and molecular clock estimation. Philosophical Transactions of the Royal Society of London. Series B: Biology Sciences 361, 1039–1054.

Rokas A, Kruger D, Carroll SB (2005) Animal evolution and the molecular signature of radiations compressed in time. Science 1933–38.

Rothman D, Hayes J, Summons R (2003) Dynamics of the Neoproterozoic carbon cycle. Proceedings of the National Academy of Sciences of the USA 100, 8124–3129.

Rumpho ME, Summer EJ, Manhart JR (2000) Solar-powered sea slugs. Mollusc/algal chloroplast symbiosis. Plant Physiology 123, 29–38.

Runnegar B (1982a) A molecular-clock date for the origin of the animal phyla. Lethaia 15, 199–205.

Runnegar B (1982b) Oxygen requirements, biology, and phylogenetic significance of the late Precambrian worm Dickinsonia, and the evolution of the burrowing habit. Alcheringa 6, 223–239.

Runnegar B (1991) Precambrian oxygen levels estimated from the biochemistry and physiology of early eukaryotes. Palaeogeography Palaeoclimatology Palaeoecology 97, 97–111.

Runnegar B (2000) Loophole for snowball Earth. Nature 405, 403–404.

Ryan JF, Mazza ME, Pang K, Matus DQ, Baxevanis AD, Martindale MQ, Finnerty JR (2007) Pre-Bilaterian origins of the Hox cluster and the Hox code: evidence from the sea anemone, Nematostella vectensis. PLoS ONE 2, e153.

Schidlowski M (1988) A 3800-million-year isotopic record of life from carbon in sedimentary rocks. Nature 333, 313–318.

Schmidt HA, Strimmer K, Vingron M, von Haeseler A (2002) TREE-PUZZLE: maximum likelihood phylogenetic analysis using quartets and parallel computing. Bioinformatics 18, 502–504.

Schwartzman D, Volk T (1991) Enhancement of weathering and the habitability of Earth. Nature 340, 457–460.

Seilacher A (1989) Vendozoa: organismic construction in the Proterozoic biosphere. Lethaia 17, 229–239.

Seilacher A, Bose PK, Pfluger F (1998) Triploblastic animals more than 1 billion years ago: trace fossil evidence from India. Science 282, 80–83.

Seilacher A, Bose PK, Pfluger F (1999) Evaluating evidence of ancient animals (reply). Science 284, 1235.

Seilacher A, Buatois LA, Mángano MG (2005) Trace fossils in the Ediacaran-Cambrian transition: behavioral diversification, ecological turnover and environmental shift. Palaeogeography Palaeoclimatology Palaeoecology 227, 323–356.

Shen Y, Knoll A, Walter M (2003) Evidence for low sulphate and anoxia in a mid-Proterozoic marine basin. Nature 423, 632–635.

Shields GA (2005) Neoproterozoic cap carbonates: a critical appraisal of existing models and the plumeworld hypothesis. Terra Nova 17, 299–310.

Shu D-G, Conway Morris S, Han J, Li Y, Zhang X-L, Hua H, Zhang Z-F, Liu J-N, Guo J-F, Yao Y, others (2006) Lower Cambrian Vendobionts from China and early diploblast evolution. Science 312, 731–734.

Siveter DJ, Williams M, Waloszek D (2001) A phosphatocopid crustacean with appendages from the lower Cambrian. Science 293, 479–481.

Sly BJ, Snoke MS, Raff RA (2003) Who came first – larvae or adults? Origins of bilaterian metazoan larvae. International Journal of Developmental Biology 47, 623–632.

Stanley S (1976) Ideas on the timing of metazoan diversification. Paleobiology 2, 209–219.

Steiner M, Reitner J (2001) Evidence of organic structures in Ediacara-type fossils and associated microbial mats. Geology 29, 1119–1122.

Summons RE, Bradley AS, Jahnke LL, Waldbauer JR (2006) Steroids, triterpenoids and molecular oxygen. Philosophical Transactions of the Royal Society of London. Series B: Biology Sciences 361, 951–968.

Svensson ME (2004) Homology and homocracy revisited: gene expression patterns and hypotheses of homology. Developmental Genes and Evolution 214, 418–421.

Taunton AE, Welch SA, Banfield JF (2000) Microbial controls on phosphate and lanthanide distributions during granite weathering and soil formation. Chemical Geology 169, 371–382.

Torsvik TH (2003) The Rodinia jigsaw puzzle. Science 300, 1379–1381.

Vacelet J, Duport E (2004) Prey capture and digestion in the carnivorous sponge Asbestopluma hypogea (Porifera: Demospongiae). Zoomorphology 123, 179–190.

Valentine JW (2002) Prelude to the Cambrian explosion. Annual Reviews of Earth and Planetary Sciences 30, 285–306.

Valentine JW, Jablonski D, Erwin DH (1999) Fossils, molecules and embryos: new perspectives on the Cambrian explosion. Development 126, 851–859.

Volkman JK (2005) Sterols and other triterpenoids: source specificity and evolution of biosynthetic pathways. Organic Geochemistry 36, 139–159.

Waggoner B (2003) The Ediacaran biota in space and time. Integrative and Comparative Biology 43, 104–113.

Walker JCG, Hays PB, Kasting JF (1981) A negative feedback mechanism for the long-term stabilization of Earth’s surface temperature. Journal of Geophysical Research 86, 1147–1158.

Wallberg A, Thollesson M, Farris JS, Jondelius U (2004) The phylogenetic position of the comb jellies (Ctenophora) and the importance of taxonomic sampling. Cladistics 20, 558–578.

Walter MR, Rulin D, Horodyski RJ (1990) Coiled carbonaceous megafossils from the Middle Proterozoic of Jixian (Tianjin) and Montana. American Journal of Science 290-A, 133–148.

Walter MR, Veevers JJ, Calver CR, Gorjan P, Hill AC (2000) Dating the 840-544 Ma Neoproterozoic interval by isotopes of strontium, carbon and sulfer in seawater, and some interpretative models. Precambian Research 100, 371–433.

Warren SG, Brandt RE, Grenfell TC, McKay CP (2002) Snowball Earth: ice thickness on the tropical ocean. Journal of Geophysical Research 107, 1–17.

Emergence of animal life 373

© 2007 The AuthorsJournal compilation © 2007 Blackwell Publishing Ltd

Watanabe Y, Martini JE, Ohmoto H (2000) Geochemical evidence for terrestrial ecosystems 2.6 billion years ago. Nature 408, 574–578.

Watanabe Y, Stewart BW, Ohmoto H (2004) Organic- and carbonate-rich soil formation similar to 2.6 billion years ago at Schagen, East Transvaal district, South Africa. Geochimica et Cosmochimica Acta 68, 2129–2151.

Welch JJ, Fontanillas E, Bromham L (2005) Molecular dates for the ‘Cambrian explosion’: the influence of prior assumptions. Systematic Biology 54, 672–678.

Welch SA, Taunton AE, Banfield JF (2002) Effect of microorganisms and microbial metabolites on apatite dissolution. Geomicrobiology Journal 19, 343–367.

Wiens M, Krasko A, Perovic S, Müller WE (2003a) Caspase-mediated apoptosis in sponges: cloning and function of the phylogenetic oldest apoptotic proteases from Metazoa. Biochimica et Biophysica Acta 17, 179–189.

Wiens M, Mangoni A, D’Esposito M, Fattorusso E, Korchagina N, Schroder HC, Grebenjuk VA, Krasko A, Batel R, Muller IM, others (2003b) The molecular basis for the evolution of the metazoan bodyplan: extracellular matrix-mediated morphogenesis in marine demosponges. Journal of Molecular Evolution 57, S60–S75.

Wiens M, Müller WEG (2006) Cell death in Porifera: molecular players in the game of apoptotic cell death in living fossils. Canadian Journal of Zoology 84, 307–321.

Wood RA, Grotzinger JP, Dickson JAD (2002) Proterozoic modular biomineralized metazoan from the Nama Group, Namibia. Science 296, 2383–2386.

Wray GA, Levinton JS, Shapiro LH (1996) Molecular evidence for deep Precambrian divergences among metazoan phyla. Science 274, 568–573.

Xiao S (2002) Mitotic topologies and mechanics of Neoproterozoic algae and animal embryos. Paleobiology 28, 244–250.

Xiao S, Dong L (2006) On the morphological and ecological history of Proterozoic macroalgae. In Neoproterozoic Geobiology and Paleobiology (eds Xiao S, Kaufman AJ). Springer, Dordrecht, the Netherlands, pp. 57–90.

Xiao S, Yuan X, Knoll AH (2000) Eumetazoan fossils in terminal Proterozoic phosphorites. Proceedings of the National Academy of Sciences of the USA 97, 13864–13689.

Yin L, Xiao S, Yuan X (2001) New observations on spicule-like structures form Doushantuo phosphorites at Weng’an, Guizhou Province. Chinese Science Bulletin 46, 1828–1832.

Yin L, Zhu M, Knoll AH, Yuan X, Zhang J, Hu J (2007) Doushantuo embryos preserved inside diapause egg cysts. Nature 446, 661–663.

Yuan X, Xiao S, Taylor TN (2005) Lichen-like symbiosis 600 million years ago. Science 5724, 1017–1020.

Zerkle AL, House CH, Cox RP, Canfield DE (2006) Metal limitation of cyanobacterial N2 fixation and implications for the Precambrian nitrogen cycle. Geobiology 4, 285–297.

Zhang SH, Jiang GQ, Zhang JM, Song B, Kennedy MJ, Christie-Blick N (2005) U-Pb sensitive high-resolution ion microprobe ages from the Doushantuo Formation in south China: constraints on late Neoproterozoic glaciations. Geology 33, 473–476.

Zhang Y, Yuan X, Yin L (1998) Interpreting late Precambrian microfossils. Science 282, 1783–1784.

Zhou C, Tucker R, Xiao S, Peng Z, Yuan X, Chen Z (2004) New constraints on the ages of Neoproterozoic glaciations in South China. Geology 32, 437–440.

Zrzavy J, Mihulka S, Kepka P, Bezdek A, Tietz D (1998) Phylogeny of the metazoa based on morphological and 18S ribosomal DNA evidence. Cladistics 14, 249–285.

Zuckerkandl E, Pauling L (1965) Evolutionary divergence and convergence in proteins. In Evolving Genes and Proteins (eds Bryson V, Vogel HJ). Academic Press, New York, pp. 97–166.